Circulating levels of cytochrome c after resuscitation from cardiac arrest: a marker of mitochondrial injury and predictor of survival

Abstract

Ca(2+) overload and reactive oxygen species can injure mitochondria during ischemia and reperfusion. We hypothesized that mitochondrial injury occurs during cardiac resuscitation, causing release of cytochrome c to the cytosol and bloodstream while activating apoptotic pathways. Plasma cytochrome c was measured using reverse-phase HPLC and Western immunoblotting in rats subjected to 4 or 8 min of untreated ventricular fibrillation and 8 min of closed-chest resuscitation followed by 240 min of postresuscitation hemodynamic observation. A sham group served as control. Plasma cytochrome c rose progressively to levels 10-fold higher than in sham rats 240 min after resuscitation (P < 0.01), despite reversal of whole body ischemia (decreases in arterial lactate). Cytochrome c levels were inversely correlated with left ventricular stroke work (r = -0.40, P = 0.02). Western immunoblotting of left ventricular tissue demonstrated increased levels of 17-kDa cleaved caspase-3 fragments in the cytosol. Plasma cytochrome c was then serially measured in 12 resuscitated rats until the rat died or cytochrome c returned to baseline. In three survivors, cytochrome c rose slightly to <or=2 microg/ml and returned to baseline within 96 h. In nine nonsurvivors, cytochrome c rose progressively to significantly higher maximal levels [4.6 (SD 2.0) vs. 1.6 (SD 0.3) microg/ml, P = 0.029] and at faster rates [0.7 (SD 0.5) vs. 0.1 (SD 0.1) microg.ml(-1).h(-1), P = 0.046] than in survivors. Plasma cytochrome c may represent a novel in vivo marker of mitochondrial injury after resuscitation from cardiac arrest that relates inversely with survival outcome.

Figures

Figure 1

Plasma cytochrome c measured by HPLC. A: Standard curve using rat heart cytochrome c dissolved in plasma obtained from pentobarbital-anesthetized rats not subject to surgical manipulation. Absorbance was measured at 393 nm (ambient temperature) and the peak area determined by integrating the beginning to the end of the chromatographic peak center at ≈ 9.5 minutes retention time (arrow in panels B and C). Solid line represents linear fit of the data (r2 = 0.9996). B: Chromatographs of plasma obtained before and after right atrial injection of cytochrome c (5 mg/kg) during spontaneous circulation in a 500 g rat demonstrating capability for detecting plasma cytochrome c. C: Representative chromatographs of plasma obtained in a 506 g rat resuscitated after 8 minutes of untreated VF. The rat died at 8.5 hours post-resuscitation (PR) preceded by progressive rise in plasma cytochrome c.

Figure 2

Hemodynamic and left ventricular measurements at baseline and post-resuscitation in rats randomized to untreated VF lasting 4 minutes (closed grey circles, n = 4), 8 minutes (closed black circles, n = 4), or to sham intervention (open circles, n = 4). Numbers in brackets indicate rats still alive. BL = Baseline; LVSWI = Left ventricle stroke work index; CI = Cardiac index; MAP = Mean aortic pressure. The shaded horizontal bar across each graph represents the 95% confidence interval of the aggregate baseline values. Mean ± SEM. *p < 0.05; †p < 0.01, and ‡p < 0.001 vs sham by one-way ANOVA and Holm-Sidak’s test for multiple comparisons.

Figure 3

Plasma cytochrome c (HPLC), lactate, and bicarbonate (HCO3−) at baseline, chest compression (for cytochrome c), and post-resuscitation in rats randomized to untreated VF lasting 4 minutes (closed grey circles, n = 4), 8 minutes (closed black circles, n = 4), or to sham intervention (open circles, n = 4). Numbers in brackets indicate rats still alive. BL = Baseline; CC = Chest compression. The shaded horizontal bar across each graph represents the 95% confidence interval of the aggregate baseline values. Mean ± SEM. *p < 0.05; †p < 0.01 and ‡p < 0.001 vs sham by one-way ANOVA and Holm-Sidak’s test for multiple comparisons.

Figure 4

Western immunoblots of plasma cytochrome c using rabbit polyclonal anti-cytochrome c primary antibody (1:2000 dilution) and goat polyclonal anti-rabbit IgG HRP conjugated secondary antibody (1:1000 dilution) in three representative experiments. Eight μl of plasma were loaded in each lane. Plasma cytochrome c levels remained unchanged in sham controls but progressively rose in rats subjected to VF. Negative control (nc) was obtained from a cytochrome c positive rat plasma (240 minutes post-resuscitation) in which the primary antibody was omitted. Positive control (pc) represents the antibody reactivity to rat heart cytochrome c from Sigma.

Figure 5

A: Western immunoblots of prohibitin (30 kDa) and cytochrome c (14 kDa) in left ventricular cytosolic (c) and mitochondrial (m) fractions (20 μg each) from 2 rats each randomized to untreated VF lasting 4 minutes (VF 4-mins), 8 minutes (VF 8-mins), or to sham intervention. Prohibitin served as loading control for the mitochondrial fraction and to exclude mitochondrial contamination of the cytosolic fraction. β-actin (42 kDa) served as loading control for the cytosolic fractions. Negative controls (nc) obtained by omitting the primary antibody. B: Western immunoblots of caspase-3, full-length (35 kDa) and cleaved fragments (19 and 17 kDa) in left ventricular cytosolic fraction (100 μg) from 2 rats each randomized to untreated VF lasting 4 minutes, 8 minutes, or to sham intervention. β-actin (42 kDa) served as loading control. Negative controls (nc) obtained by omitting the primary antibody. Longer exposure times were required to detect the cleaved fragments.

Figure 6

Densitometry of immunoblots demonstrating numerical increases in mitochondrial cytochrome c relative to prohibitin and cytosolic cytochrome c relative to β-actin and statistically significant increases in 17 kDa cleaved caspase-3 fragments in the cytosolic fraction relative to pro-caspase-3 and β-actin at 240 minutes post-resuscitation. Rats were randomized to untreated VF lasting 4 minutes (closed grey bars, n = 4), 8 minutes (closed black bars, n = 4), or to sham intervention (open bars, n = 4). Mean ± SEM. *p < 0.05 vs sham by one-way ANOVA and Dunn’s test for multiple comparisons.

Figure 7

Serial measurements of plasma cytochrome c by reverse-phase HPLC in rats successfully resuscitated after 8 minutes of untreated VF. Measurements were made until the cytochrome c level had returned to baseline or the rat had died. Open circles represent survivors (n = 3) and closed circles represent non-survivors (n = 9).

Figure 8

Detection of apoptosis in rat blood leukocytes by annexin V-FITC and propidium iodide labeling followed by flow cytometry. Flow scattergrams of total leukocytes, neutrophils, and lymphocytes from blood obtained from representative rats at baseline in one animal and at post-resuscitation (PR) after 8 minutes of untreated VF in another. The x- and y-axes represent relative fluorescence intensity. The percentages represent number of events within the respective quadrants.