An investigation into the relationships between telomere length and cardiac function and geometry
Date
2014-09-09
Authors
Raymond, Adrew Ryan
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Telomere length is inversely associated with the presence of heart failure. However, it is uncertain whether a reduced telomere length has pathophysiological consequences in the heart or is merely an epiphenomenon. The relationship between telomere length and heart failure may be causal, or it may reflect an association with inflammation or with risk factors for coronary artery disease, in which case it would be specific to ischaemic heart disease. Therefore, the main aim of this thesis was to determine whether telomere attrition is a cause of heart failure or a consequence of ongoing processes (eg. cardiomyocyte apoptosis, senescence, or mitochondrial dysfunction) in heart failure. Hence, I investigated the relationships between telomere length and a non-ischaemic form of heart disease (excludes oxidative stress and inflammation associated with coronary artery disease and its risk factors) and between telomere length and measurements of cardiac remodelling and systolic function prior to the development of heart failure.
Hence, in my first study I evaluated associations between leukocyte telomere length and the presence or severity of a non-ischaemic form of heart failure, namely idiopathic dilated cardiomyopathy (IDC). Relative telomere length (log T/S), measured using a quantitative real-time polymerase chain reaction assay, was correlated with age in patients with IDC (p=0.002, n=223) and in control participants (p=0.023, n=227). However, no differences in leukocyte log T/S were noted between IDC patients and healthy controls both prior to (p=0.16) and after (log T/S: IDC=-0.22±0.022 vs. control=-0.22±0.024, p=0.114) adjustments for confounders. Moreover, with adjustments for age, sex, regular alcohol consumption and smoking, log T/S was not associated with echocardiographic left ventricular (LV) ejection fraction (EF) or LV end diastolic diameter (EDD) in patients with IDC. In conclusion, relative leukocyte telomere length is not associated with the presence of IDC or the severity of heart failure in IDC (as indexed by LVEF and LVEDD). These data suggest
that telomere attrition is unlikely to be a cause of IDC (non-ischaemic heart disease) or a consequence of heart failure in general.
To further explore whether telomere shortening is a cause of cardiac remodelling and systolic dysfunction, I investigated the relationships between telomere length and an animal model of alcoholic cardiomyopathy (a form of dilated cardiomyopathy). Sprague-Dawley (SD) rats received either drinking water with (ethanol: n=19) or without (control: n=19) 5% (v/v) ethanol ad libitum, for four months. Echocardiography was performed to determine LV dimensions and function in vivo and isolated perfused heart preparations (ethanol n=10, control n=10) were performed to obtain ex vivo load-independent measurements of LV chamber geometry and function. Ethanol consumption resulted in LV dilatation (LVEDD: ethanol=8.20±0.14mm vs. control=7.56±0.11mm, p=0.0014; the volume intercept at 0mmHg (V0) of the LV end diastolic pressure-volume relationship: ethanol=0.40±0.03ml vs. control=0.31±0.02ml, p=0.020), and a reduced LV relative wall thickness (ethanol=0.50±0.08mm vs. control=0.60±0.09mm, p=0.0011); but no changes in LV systolic chamber function as indexed by endocardial fractional shortening (FSend) or the slope of the LV systolic pressure-volume relation (end systolic elastance-Ees). Ethanol administration resulted in a marked increase in cardiomyocyte apoptosis (ethanol=0.85±0.13% vs. control=0.36±0.06%, p=0.0021), and % apoptosis was correlated with LVEDD (r=0.39, n=38, p=0.021) and V0 (r=0.44, n=19, p=0.046). However, no differences in leukocyte, cardiac, liver, or kidney telomere length were noted between ethanol and control groups; and cardiac telomere length was not associated with indices of LV dilatation or apoptosis. In conclusion, chronic alcohol-induced cardiac remodelling and apoptosis in the absence of systolic dysfunction are independent of changes in telomere length.
Having shown in two studies that telomere attrition is unlikely to be a pathophysiological cause of heart failure, I then explored whether telomere shortening in
heart failure is due to processes such as β-adrenergic receptor activation and inflammation, which occur in heart failure. I first evaluated whether LV dilatation and systolic chamber dysfunction produced by chronic β-adrenergic receptor (β-AR) activation are associated with leukocyte or cardiac telomere shortening. Six months of daily injections of the β-AR agonist, isoproterenol (0.02 mg.kg-1.day-1) (ISO: n=16), or the saline vehicle (control: n=15) to SD rats, resulted in LV dilatation (LVEDD: ISO=9.18±0.60mm vs. control=8.41±0.90mm, p=0.010; V0: ISO=0.45±0.06ml vs. control=0.37±0.08ml, p=0.03) and LV systolic chamber dysfunction (FSend: ISO=39.88±5.18% vs. control=45.14±5.64%, p=0.01; Ees: ISO=608.60±175.40 vs. control=901.10±229.60, p=0.01). Although leukocyte log T/S decreased over six months in rats receiving either the β-AR agonist or the saline vehicle (p<0.05); neither cardiac (ISO=-0.10±0.14 vs. control=-0.15±0.12, p=0.35) nor leukocyte log T/S (ISO=-0.11±0.19 vs. control=-0.15±0.18, p=0.52) were altered by chronic β-AR activation. Moreover, cardiac telomere length was not associated with indices of LV dilatation or systolic chamber function. In conclusion, chronic β-AR activation that is sufficient to produce LV dilatation and systolic chamber dysfunction was not associated with alterations in leukocyte or cardiac telomere length. Hence, telomere shortening in chronic heart failure is unlikely to be attributed to chronic β-AR activation.
I then evaluated the impact of chronic inflammation on telomere length, LV remodelling and systolic function in an animal model. After 12 months of fortnightly intraperitoneal injections of 2.5x107 Staphylococcus aureus cell walls (S. aureus: n=30) or saline vehicle (control: n=30) in SD rats, there were no differences in LV dimensions (EDD, V0) or systolic chamber function (FSend, Ees) between the two groups. In the control group no differences in plasma C-reactive protein concentrations (a marker of inflammation) were noted at 3 or 11 months compared to the baseline concentrations at 1 month. However, in the S. aureus group there was an increase in CRP concentrations at both 3 months (S. aureus: 3
months=435.40±89.95μl.ml-1 vs. 1 month=347.80±101.90μl.ml-1, p<0.0001) and 11 months (S. aureus: 11 months=511.40±93.34μl.ml-1 vs. 1 month=347.80±101.90μl.ml-1, p<0.0001), and CRP concentrations were increased in the S. aureus group compared to the control group at 11 months (S. aureus=511.40±93.30μl.ml-1 vs. control=391.20±122.10μl.ml-1, p<0.01). Leukocyte log T/S (at 11 months) (S. aureus =-0.21±0.15 vs. control=-0.053±0.11, p = 0.002) and cardiac log T/S (S. aureus=-0.20±0.10 vs. control=-0.11±0.084, p= 0.0021) were reduced in the S. aureus group. However cardiac log T/S was not associated with LVEDD, V0, FSend, or Ees. In conclusion, a chronic inflammatory stimulus that produced reductions in leukocyte and cardiac telomere length did not induce LV dilatation and a reduction in systolic chamber function. Hence, the negative association between telomere length and heart failure may be due to the impact of inflammatory processes on telomere length rather than signifying a cause of heart failure.
Having shown that telomere length may be reduced as an epiphenomenon due to ongoing inflammatory processes with no pathophysiological consequences on cardiac geometry or function in rats, I attempted to validate these findings in humans. Hence, I evaluated the relationship between leukocyte telomere length, indices of inflammation, LV remodelling, and systolic function in a cross-sectional human study. Leukocyte telomere length was determined in 622 participants from a randomly selected community based cohort. There was a negative bivariate relationship between T/S ratio and age in the group (p<0.0001) and telomere length was reduced in participants that regularly consumed alcohol compared to those who did not (p=0.01). No relationships between log T/S and smoking, sex, BMI, or blood pressure were noted. However, there was a negative relationship between leukocyte telomere length and C-reactive protein concentrations after multivariate adjustments for confounders (r=-0.11; p=0.0039; n=662). There were no associations between log T/S and echocardiographic indices of LV dilatation (LVEDD) or systolic
dysfunction (FSend, FSmid) in the participants. In conclusion, leukocyte telomere length was inversely associated with indices of inflammation. However, leukocyte telomere length was not associated with indices of LV remodelling and systolic function. These findings are consistent with the previous findings that chronic inflammation in rats capable of producing a marked decrease in cardiac telomere length has no pathophysiological consequence on cardiac dimensions or systolic function. Hence, inflammation may be an important determinant of leukocyte telomere length in humans, but leukocyte telomere length is not a bio-marker of adverse cardiac geometry or function.
The main findings from my thesis are as follows; a reduced telomere length was not associated with IDC (a non-ischaemic form of heart disease) nor with a model of alcoholic cardiomyopathy, which demonstrates that reduced telomere length may not be associated with all types of cardiomyopathy. Although chronic β-AR activation and ethanol consumption induced cardiac remodelling and β-AR activation induced systolic dysfunction, cardiac telomere length was not altered. Moreover, in IDC, alcoholic cardiomyopathy, chronic β-AR stimulation, chronic inflammation, and in a random, community-based population a reduced telomere length was not associated with features of heart failure such as LV dilatation (LVEDD and V0), reduced systolic chamber function (FSend and Ees), reduced intrinsic myocardial function (FSmid and Een), and myocardial apoptosis. However, chronic inflammation resulted in reductions in leukocyte and cardiac telomere length in the absence of changes in cardiac geometry and systolic function. These data indicate that the pathophysiological processes of heart failure, namely LV dilatation and reduced systolic function, may be independent of changes in telomere length and that telomere length may only be reduced as an epiphenomenon due to inflammatory processes in heart failure.