Author + information
- Received January 23, 2001
- Revision received July 10, 2001
- Accepted August 9, 2001
- Published online November 15, 2001.
- Chi-Woon Kong, MD*,
- Tai-Ger Hsu, PhD†,
- Fung-Jou Lu, PhD‡,
- Wan-Leong Chan, MD, FACC* and
- Kelvin Tsai, MD*,* ()
- ↵*Reprint requests and correspondence: Dr. Kelvin Tsai, Oxidative Stress Clinical Research Group and Division of Critical Care, Department of Medicine, Taipei Veterans General Hospital, 2F, 201, Section 2, Shih-Pai Rd., Taipei, 112, Taiwan
The purpose of this study was to examine the changes in leukocyte mitochondrial transmembrane potential (MTP) and its association with apoptosis in congestive heart failure (CHF).
Congestive heart failure is a heterogeneous syndrome with multiple hemodynamic, neuroendocrine and immune abnormalities. Although edematous CHF may be associated with endotoxemia and increased cytokine production, peripheral blood leukocyte functions in advanced CHF remain unclear.
Thirty patients with acute decompensated CHF (mean age [± SEM] 74.9 ± 3.1 years) and 20 healthy controls underwent determination of MTP, intracellular oxidants and apoptosis in three subsets of peripheral blood leukocytes. The measurements were repeated after the time of recompensation.
Patients with acute CHF showed marked MTP reduction and increased intracellular oxidant formation in three subsets of leukocytes upon entry into the study. These changes were more prominent in patients with peripheral edema. The decline in MTP was correlated with the severity of the peripheral edema and plasma concentration of cortisol, nitrogen metabolites and tumor necrosis factor-alpha (p < 0.01). After clinical stabilization, MTP gradually recovered. Leukocytes underwent increased propensity of apoptosis one week after the time of recompensation.
The mitochondrial depolarization and apoptosis of leukocytes in decompensated heart failure suggest that CHF is associated with severity-dependent impairments in leukocyte function. Accentuated hormonal and cytokine abnormalities and increased circulating oxidants may contribute to these changes. Early and aggressive management of advanced heart failure is helpful in the recovery of these immune abnormalities.
Congestive heart failure (CHF) is associated with altered immune status and increased oxidative stress (1,2). Increased levels of inflammatory cytokines and neopterin have been demonstrated in edematous or cachectic CHF patients (3–5). Several recent researches proposed the paradigm that intestinal bacterial translocation, a phenomenon secondary to mesenteric venous congestion and intestinal edema, may induce endotoxemia and lead to immune activation in CHF patients (6). Of note, CHF is associated with more heterogeneous neuroendocrine and immune disturbances (7,8). For example, an increased level of blood cortisol, which may potentially cast deleterious effects on the physiological state and vitality of leukocytes (9,10), is a part of CHF-associated metabolic alternations (3). Furthermore, the impairment in leukocyte oxidative bursts in patients with severe heart failure seems to further support the “negative facet” of CHF-related immune disturbances (5).
Cellular energization status is crucial to the activity and vitality of cells (11). Mitochondrial transmembrane potential (MTP), the driving force of cellular adenosine triphosphate (ATP) formation, is an important determinant of the energization status and physiological activity of the cell, and also constitutes an obligate step in cell-death programs (12–15). A reduction in MTP can be triggered by various apoptotic inducers such as glucocorticoids, ischemia, reactive oxygen species (16), tumor necrosis factor-alpha (TNF-α) (17)and nitrogen metabolites (18). These inducers may cause functional disturbance in mitochondrial respiratory chain components or the inappropriate opening of the permeability transition (PT) pores, resulting in mitochondrial depolarization and MTP disruption (13). The mitochondrial functional status of immunocompetent cells in patients with CHF has not been investigated yet.
Because the extent of neuroendocrine and immune disturbances in CHF patients is correlated with the severity of heart failure and hemodynamic parameters (2,7,19), it is tempting to speculate that these abnormalities may be more prominent at the deterioration stage of CHF. To gain further insight into the mixed and complex responses of the immune system to CHF, we conducted a prospective, observational study to investigate peripheral blood leukocyte MTP and its propensity for apoptosis in 30 decompensated CHF patients. The study subjects were grouped according to whether peripheral edema existed or not, and were sequentially gauged to evaluate time-sequence changes after appropriate management of CHF.
From October 1999 to October 2000 we prospectively enrolled 40 patients who were admitted to the intensive care unit of the Taipei Veterans General Hospital because of acute decompensated CHF. The patients were recruited within 6 h after onset of their symptoms. The diagnosis of CHF was based on the presence of dyspnea, cardiomegaly, and documented left-ventricular dysfunction (all patients had a left-ventricular ejection fraction measured by radionuclide ventriculography of <40%). Patients with premorbid inflammatory or infectious disorders, malignancy, uremia, and decompensated liver disease, and those taking anti-inflammatory agents, steroid hormones, or antioxidants prior to entering this study, were excluded. Patients developing cardiogenic shock or those mandating nitrite therapy during hospitalization were also excluded due to the possible interference in leukocyte mitochondrial function (9). Recompensation of CHF was defined as the disappearance of dyspnea at rest, peripheral edema, hypoxemia and pulmonary infiltrations in chest radiographs. Twenty healthy volunteers free of CHF, or any condition compatible with the excluding criteria listed above, donated blood samples as reference materials after overnight fasting. Written informed consent was obtained from all study participants, and the performance of this study was approved by the Research Ethics Committee at the Taipei Veterans General Hospital, Taiwan.
Twenty of the 35 included patients (8 in the endematous group and 12 in the nonedematous group) underwent pulmonary artery cannulation with a balloon-tipped, flow-directed, 7F Swan-Ganz catheter through the internal jugular or subclavian vein. Pressure waveforms were recorded in the right atrium, pulmonary trunk and pulmonary artery wedge positions. Cardiac output was determined by applying the thermo-dilution method.
Immediately upon fulfillment of the inclusion criteria, 10 ml of venous blood was obtained by venipuncture for sampling. Part of the blood (5 ml) was anticoagulated with heparin and assayed immediately for various fluorescence stainings (see subsequent text). The remaining blood samples were EDTA-anticoagulated, centrifuged and stored in a frozen state at −70°C. Repeated blood sampling was performed under fasting states at the time of hemodynamic recompensation (mean duration 5.6 ± 2.4 days after admission) and one week thereafter.
The heparinized blood was divided into several 200 μl aliquots. Each aliquot was treated with 3 ml of 1:10 erythrocyte lysing buffer (PharMingen, San Diego, California) for 10 min. The supernatant was discarded and the cells were washed with phosphate-buffered saline and resuspended with Hank’s balanced salt solution (Gibco BRL, Paisley, Scotland) to approximately 105cells/ml. The viability of leukocytes was confirmed to be more than 95% by trypan blue exclusion.
For assaying DNA fragmentation, polymorphonuclear leukocytes (PMN) and peripheral blood mononuclear cells (PBMC) were separated from whole blood by Histopaque-1119 (Sigma, St. Louis, Missouri) and Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) centrifugation. After isolation and resuspension, the cells were fixed with 1% paraformaldehyde and 70% ethanol and stored under −70°C for further assay.
Aliquots of leukocytes were labeled separately with the following fluorescent probes. The MTP was specified by incorporating the fluorochrome 5,5′,6,6′-tetrachloro-1,1′3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1) (Molecular Probes, Eugene, Oregon) in cell staining. The leukocyte suspension was incubated with 5 μmol/l JC-1 for 20 min at 37°C. After staining, JC-1 was incorporated into the mitochondria, where it either formed monomers (green fluorescence, 527 nm) or, at high MTP, aggregates (red fluorescence, 590 nm). The ratio between fluorescence intensity of JC-1 aggregates and monomers can reliably reflect MTP (20). In addition, intracellular oxidants were evaluated by incubating the cells with 20 μmol/l 2′,7′-dichlorofluorescein (DCF) diacetate (Molecular Probes) for 20 min at 37°C (21).
The DNA fragmentation assay
The DNA fragmentation of nuclei was assessed using terminal deoxynucleotidyltransferase to incorporate fluorescein-isothiocyanate-dUTP into nuclei (TUNEL assay), following the manufacturer’s protocol (APO-BRDU Kit, PharMingen, San Diego, California) (22). The percentage of apoptosis was calculated as the number of cells in the high fluorescence intensity population divided by the total number of cells (PMN or PBMC) analyzed.
Fluorescence was analyzed by cytometry using a FACScan (Becton Dickinson, San Jose, California) fitted with an air-cooled argon laser emitting at 488 nm. During analysis, a gate was set on the dot plot of forward and side scatter to include PMN, monocytes or lymphocytes. The identity of cell populations in each analysis was confirmed by counterstaining with CD45 and CD14 antibody reagents (23). The total number of events from each sample was made such that at least 5,000 events were collected for each cell population. To ensure consistency of data among the different measurements, the photomultiplier values of the detector in FL1 and FL2 remained constant and were set at 450 V throughout all the experiments. All the analyzing procedures were undertaken within 3 h after blood sampling.
Associating plasma factors
For the following assays, EDTA-anticoagulated plasma samples were used. Nitrogen metabolites, including nitrites and nitrates, were measured by a modified Griess method (BIOXYTECH Nitric Oxide Non-Enzymatic Assay, Oxis, Portland, Oregon), with a detection sensitivity of 1 μmol/l. Lipid peroxides (including malondialdehyde and 4-hydroxyalkenals) were analyzed by a colorimetric assay kit (LPO assay kit, CalBiochem, San Diego, California), with a detection limit of 0.1 μmol/l. Cortisol was assayed by a radioimmunoassay (RIA) kit (DSL-2000 Cortisol RIA kit, Diagnostic Systems, Webster, Texas), with a detection sensitivity of 2.7 nmol/l. Both TNF-α and interleukin-6 (IL-6) concentrations were measured using ultrasensitive ELISA kits (BioSource, Camarillo, California), with a minimum detectable range of 1.7 pg/ml and 0.7 pg/ml, respectively.
All continuous data were expressed as mean ± SEM. Time-sequence changes and group comparisons were assessed by the Mann-Whitney Utest or two-way analysis of variance (ANOVA) models with repeated-measures ANOVA methods when appropriate. Spearman rank correlation or Pearson’s product moment correlation analyses were used to establish the relation between variables. A p value < 0.01 was considered statistically significant.
Of the 40 CHF patients initially recruited, 5 mandating nitrate therapy were excluded, as well as another 5 who developed infectious complications or eventually died during the hospitalization period. The remaining 30 patients who achieved recompensation for a mean of 5.3 ± 1.2 days were included in the final assessments. Among them, 15 presenting with peripheral edema were categorized into the edematous group, with the remaining 15 categorized into the nonedematous group. Peripheral edema was graded according to its severity and range of distribution (0, no edema; 1, mild; 2, moderate; 3, severe). The clinical characteristics of these patients are shown in Table 1.
Analysis by flow cytometry revealed a significant reduction of MTP in PMN, monocytes, and lymphocytes at the onset of decompensated CHF (Fig. 1, left panel). The decline of MTP was more prominent in the edematous than in the nonedematous patients, while the difference being statistically significant only in lymphocytes. After clinical stabilization of CHF, MTP gradually increased, reaching normal or subnormal levels one week after recompensation. The sequential changes in MTP in the three subsets of leukocytes were all statistically significant.
Intracellular oxidants were significantly increased in the three subsets of leukocytes upon patients’ admissions (Fig. 1, right panel). The highest levels were noted for monocytes. Patients with edematous CHF exhibited higher DCF fluorescence levels than did those with nonedematous CHF. The DCF fluorescence gradually declined after recompensation of CHF, but remained slightly supranormal one week later.
The DNA fragmentation assay revealed a delayed appearance of apoptotic leukocytes in CHF subjects one week after the time of recompensation (Fig. 2). The percentage of apoptotic cells was higher in PMN (up to 42.9%) than in PBMC (up to 20.8%). Leukocytes from patients with edematous CHF showed a higher percentage of apoptosis than those of nonedematous subjects, but the difference did not reach statistical significance except that at one week for PBMC (p < 0.01 by Mann-Whitney Utest).
Associating plasma factors
Figure 3displays time-sequence changes of plasma nitrogen metabolites, lipid peroxides, cortisol and inflammatory cytokine (TNF-α and IL-6) concentrations during the observation period. Plasma nitrite/nitrate levels increased approximately two-fold from admission, as compared with the control value. More prominent changes were noted in plasma lipid peroxide levels, with the increase reaching sevenfold compared with the control value. The increases in plasma oxidants were more pronounced in the edematous subjects. Plasma cortisol levels increased significantly from admission in the edematous CHF subjects, and the levels normalized one week after the time of recompensation. Finally, plasma TNF-α and IL-6 levels markedly increased in patients with edematous CHF, and remained supranormal throughout the study period.
Correlations of MTP with clinical indexes
The levels of MTP on admission are variably correlated with clinical and laboratory markers (Table 2). When the patients were considered collectively, MTP levels declined with increased severity of peripheral edema and higher right atrial pressure (p < 0.01). In contrast, MTP levels did not correlate significantly with indexes of left ventricular function. The MTP was inversely related with most plasma markers, with the correlations reaching statistical significance in nitrogen metabolites, cortisol and TNF-α (p < 0.01).
The major findings of this study provide evidence showing that impaired leukocyte mitochondria functional status and an increased propensity for apoptotic death may constitute an important part of immune disturbances in CHF patients. Mitochrondria depolarization occurred in three subsets of peripheral leukocytes and was more prominent in patients with peripheral edema. The mitochondrial dysfunction gradually recovered after clinical stabilization. These findings suggest that CHF is associated with severity-dependent immune dysfunctions that are partly recoverable with appropriate management.
Possible mechanisms underlying MTP disruption
Several factors may potentially contribute to the leukocyte MTP disruption in patients with CHF. First, increased intracellular oxidants, as evidenced by increased DCF fluorescence, may lead to oxidation of carrier molecules in the respiratory chain and/or mitochondria pyridine nucleotides, which either hamper the respiratory coupling or induce mitochondrial PT (16)with resultant MTP disruption. Second, increased DCF fluorescence might also signify greater production of mitochondrial reactive nitrogen species such as nitric oxide or peroxynitrite (21), which is able to cause reversible or irreversible inhibition of respiratory enzymes and lead to MTP disruption (24–26). Alternatively, increased plasma nitrogen metabolites, probably secondary to endothelial nitric oxide synthase activation by endotoxemia and increased circulating cytokine in edematous CHF (27,28), may also trigger MTP disruption in leukocytes (9). Third, plasma TNF-α, with the potential to deplete cells of nicotinamide adenosine diphosphate (NAD) and adenosine triphosphate (ATP) due to the activation of poly-ADP-ribose polymerase (29), is another culprit in mitochondrial depolarization. Although it was unclear whether the plasma concentration of TNF-α in our CHF patients was adequate to elicit these changes, the fact that leukocytes from patients with CHF have both enhanced expression of surface TNF-α receptors and more responsiveness to TNF-α (30)seems to heighten this possibility. Finally, cortisol also seems to be a potential contributor of MTP disruption (31,32). The plasma concentration of cortisol increased to about 500 nmol/l in our CHF patients, exceeding concentrations that were able to induce lymphocyte MTP disruption and apoptosis in vitro (33).
Disruption and apoptosis of MTP
Recent evidence proposed the paradigm that the reduction in MTP may constitute an early and committed step in cell apoptosis (14,15). The dissipation in MTP may induce apoptotic programs by disturbing mitochondrial ATP production and by inducing the release of mitochondrial matrix solutes, cytochrome C, and other pro-apoptotic factors (34). In the present study, a significant percentage (23.5%) of peripheral blood PMN underwent apoptosis at the decompensated stage of CHF. Intriguingly, the propensity of PMN apoptosis increased for a period after the time of recompensation. Also, PBMC underwent apoptosis at this stage. It seems that these phenomena may reflect the temporal effect of downstream events in the apoptotic program induced by mitochondrial depolarization.
Comparisons between edematous and nonedematous CHF
The increased plasma concentrations of TNF-α and IL-6 in CHF patients, especially those with peripheral edema, corroborate previous findings (1). The in-crease in inflammatory cytokines may potentially augment phagocytic oxidant production and lead to more intravascular oxidants, as shown by Keith et al. (2)and our study (Fig. 3B). Our study further demonstrates a two-fold increase in plasma nitrogen metabolites and a 1.5-fold increase in plasma cortisol in the edematous CHF patients. Because all these factors have the potential to induce mitochondrial depolarization and apoptosis, the greater degree of their abnormalities seems to account for the more prominent leukocyte dysfunctions in the edematous CHF patients. The significant correlation of MTP levels with the severity of peripheral edema renders further support to this notion.
Possible clinical implications
The CHF-related immune alterations seem to be multifold and remain to be clarified. Congestive heart failure with peripheral edema or cachexia has been shown to be associated with immune activation and increased circulating pro-inflammatory cytokines (1,35). Conversely, reduced host defenses, as evidenced by an increased susceptibility to pneumonia, have also been reported in CHF patients (36). The present study further explores the pathomechanisms underlying the immune disturbances associated with CHF by demonstrating an impaired leukocyte mitochondria functional status and an increased propensity for apoptosis in severe and untreated heart failure. These phenomena may constitute a pivotal facet of immune dysfunction in advanced CHF. That leukocyte dysfunction partly improved after the restoration of hemodynamic functions may underscore the importance of early and aggressive management of advanced heart failure.
Unlike well-controlled laboratory investigations where a single variable can be manipulated at a time, circulating leukocytes are under simultaneous influence by a number of contributing factors, making a definite conclusion very difficult to reach. In addition, medications used in the treatments of CHF, such as catecholamines, also have the potential to induce lymphocyte apoptosis (37), thus making the interpretation of this study’s results more complex. Cautions must also be taken in the interpretation of the results of the TUNEL assay, as it may be too sensitive to detect not only the specific DNA fragmentations in apoptosis, but also the nicks in DNA that were produced by topoisomerase (38)or the random DNA degradation occurring during necrosis (39,40). Further in-depth studies are demanded to verify the different pathways that are involved in the immune abnormalities associated with acute CHF.
Congestive heart failure is a state of physiological stress. The stress resides not only in the level of cardiovascular function, but also at the cell level, as exemplified by leukocyte mitochondrial depolarization and apoptosis. The impaired functional status of peripheral blood leukocytes may impair host defenses in the acute deterioration stage of CHF. Successful management of CHF not only improves the patient’s hemodynamic status but also restores the leukocyte energization status. These findings may shed new light on both the pathogenesis and the treatment of immune disturbances in CHF.
☆ This project was supported, in part, by grant VAC 90-44 from the Taipei Veterans General Hospital and by grant NSC 89-2320-B075-046 from the National Science Council, Taiwan.
- congestive heart failure
- 5,5′,6,6′-tetrachloro-1,1′3,3′-tetraethylbenzimidazolcarbocyanine iodide
- mitochondrial transmembrane potential
- peripheral blood mononuclear cells
- polymorphonuclear leukocytes
- permeability transition
- tumor necrosis factor
- Received January 23, 2001.
- Revision received July 10, 2001.
- Accepted August 9, 2001.
- American College of Cardiology
- Keith M,
- Geranmayegan A,
- Sole M.J,
- et al.
- Torre-Amione G,
- Kapadia S,
- Lee J,
- et al.
- Anker S.D,
- Chua T.P,
- Ponikowski P,
- et al.
- Wiedermann C.J,
- Beimpold H,
- Herold M,
- et al.
- Lommi J,
- Kupari M,
- Koskinen P,
- et al.
- Brunetti M,
- Martelli N,
- Colasante A,
- Piantelli M,
- Musiani P,
- Aiello F.B
- Crompton M
- Marchetti P,
- Hirsch T,
- Zamzami N,
- et al.
- Hempel S.L,
- Buettner G.R,
- O’Malley Y.Q,
- Wessels D.A,
- Flaherty D.M
- Qwens M.A,
- Loken M.W
- Fortenberry J.D,
- Owens M.L,
- Brown L.A
- Rockey D.C,
- Chung J.J
- Agarwal S,
- Drysdale B.E,
- Shin H.S
- Distelhorst C.W
- Zamzami N,
- Marchetti P,
- Castedo M,
- et al.
- Brunetti M,
- Martelli N,
- Colasante A,
- Piantelli M,
- Musiani P,
- Aiello F.B
- Froelich-Ammon S.J,
- Osheroff N