Telomere length in vascular tissues from patients with atherosclerotic disease

Abstract

Objectives

The present study was aimed at evaluating telomere length in blood and in different vascular tissues with or without atheroma, in 3 groups of subjects: a group of atherosclerotic subjects who underwent surgery (Atherosclerosis-Surgery), a second group of subjects with asymptomatic atherosclerotic carotid plaques but who did not undergo cardiovascular surgery (Atherosclerosis-No surgery), and a third group of subjects without atherosclerotic disease (Controls). The main objective was to determine if there is in vivo regulation of telomere length in situ by atherosclerotic lesions.

Methods

A total of 84 subjects (mean age 69±8 years) were studied. Blood and arterial tissue telomere lengths were determined by Southern blotting. Personal medical history (diabetes, hypertension, cardiovascular disease, dyslipidemia), family medical history, drug intake, and lifestyle were evaluated in the entire population through the use of a questionnaire.

Results and Conclusion

Arterial segments which did not develop atherosclerosis such as the saphenous vein and internal mammary artery, had longer telomere length than aortic segments. On the other hand, telomere length was shorter in aortic tissues which presented atherosclerotic lesions compared to corresponding tissues without atherosclerotic lesions. These results also suggest tissue regulation of telomere size by local factors likely related to oxidative stress responses.

Introduction

Several studies have shown decreased telomere length in white blood cells (1, 2, 3, 4, 5) and in vascular cell cultures (6, 7) from patients with atherosclerotic disease. Atherosclerosis is initiated by repeated mechanical, hemodynamic, and/or immunological injury to the vessel wall and focal regions of the endothelium leading to cellular senescence (8, 9). This phenomenon can occur independently of replicative age, following exposure to multiple types of stress (stress-induced senescence) such as mitogenic stress, DNA damage (10) and oxidative stress (11). Increased levels of reactive oxygen species (ROS) are found in all layers of the atherosclerotic arterial wall, and particularly in the plaque itself (12). Telomeres, which are particularly rich in sequences with high guanosine content, are one of the DNA structures most sensitive to oxidative damage (13). ROS can accelerate telomere loss in vitro (14) by inducing breaks in DNA strands as well as base and nucleotide modifications in telomeres (15).

Although telomere length is similar throughout the various organs in the human newborn (16), it is now generally accepted that differences observed in telomere length at a given age depend on telomere attrition which in its turn depends on both replicative cell activity and cumulative effects of oxidative stress.

There are presently no data confirming in vivo regulation of telomere length in situ by atherosclerotic lesions. The aim of the present study was thus to evaluate telomere length in blood and in various vascular tissues with or without atheroma, in patients with atherosclerotic disease.

Methods

Population

Three groups of subjects were included in the present study: Group 1 (Atherosclerosis-Surgery) was comprised of 48 subjects (36 men and 12 women) with a mean age 69 ± 8 years. Subjects had surgery for various atherosclerotic diseases (coronary diseases, degenerative aortic valvular diseases, aortic aneurysms, lower limb arteriopathy). Blood samples for DNA extraction and leukocyte telomere length (LTL) measurements were available in 18 subjects. In order to compare LTL measurements in these patients, two other age- and gender-adjusted groups of subjects were studied: one constituted of 18 subjects with asymptomatic atherosclerotic carotid plaques but who did not undergo cardiovascular surgery (Group 2, Atherosclerosis-No surgery) whereas Group 3 (Controls) was comprised of 18 subjects without atherosclerotic disease.

Through the use of a questionnaire, personal medical history (diabetes, hypertension, cardiovascular disease, dyslipidemia), family medical history, drug intake, and lifestyle were evaluated in the entire population.

Samples

Blood and vascular tissues samples, considered as surgical leftovers, were obtained from surgical interventions in the Department of Cardiac Surgery and Transplantation of the CHU in Nancy. Vascular tissues were collected in ice-cold sterile

PBS, and immersed in liquid nitrogen and stored at -80° C. Blood samples were collected in EDTA tubes. All protocols were approved by the institutional ethical committee of the CHU of Nancy.

Telomere length analysis

White blood cell DNA was extracted by salting out method (Miller et al., 1998). DNA from vascular tissue was isolated with a DNA isolation kit (Qiagen, Courtaboeuf cedex, France) according to the manufacturer's protocol. DNA samples were digested overnight with the restriction enzymes Hinf I (40 U) and Rsa I (40 U) (Roche Diagnostics GmbH, Germany). Twenty-four DNA samples (~ 3 g each) and 5 DNA ladders (1 kb DNA ladder plus DNA/Hind III fragments; Invitrogen, USA) were loaded onto a 0.8% agarose gel (15 cm x 25 cm) and separated by electrophoresis at 40V (PowerPac Basic, Biorad) for 20 h. The gels were treated (0.25 N HCl, 0.5 mol/L NaOH/1.5 mol/L NaCl and 0.5 mol/L Tris, pH 8/1.5M NaCl). The DNA was then transferred for 1.5 hours onto a positively-charged nylon membrane (Amersham Pharmacia Biotech UK Limited, England) using a vacuum blotter (Biorad). The membranes were hybridized at 65°C with the telomeric probe [digoxigenin 3‘-end labeled 5‘-(CCTAAA)₃]. The digoxigenin-labeled probe was detected by the digoxigenin luminescent detection procedure (Roche diagnostics GmbH, Germany). Each DNA sample was measured in duplicate. If the difference between the duplicates was >5%, a third measurement was performed, and the mean value of the 2 results <5% apart was taken. A calibrator sample was included on every gel to allow correction for interassay variability (6.2 %).

Statistical Analysis

Statistical analysis was carried out using NCSS software. The results are presented as the mean ± standard deviation. Parameter association was studied using univariate or multivariate analyses. A value of p < 0.05 was considered to be significant. For technical reasons, telomere length was not always available for all tissues in all patients. Thus, the number of analyzed samples is indicated in each table and figure.

Results

Population characteristics

Table 1 shows the clinical characteristics of the 48 patients who underwent cardio-vascular surgery. Patient ages ranged from 45 to 81 years, with a mean age of 69±8 years.

Table 1.

Clinical characteristics of atherosclerotic patients who underwent surgery

Description of the population	Surgery Group
Number of patients	48
Age (years)	69 ± 8
Gender
- Men	34 (71 %)
- Women	14 (29 %)
Medical history
Diabetes mellitus
- type 1	2 (4.2 %)
- type 2	11 (23 %)
Hypertension	28 (59 %)
Current or ex-smoker	24 (50 %)
Obesity	13 (27 %)
Dyslipidemia	35 (73 %)
Arterial parameters
Systolic Blood Pressure (mmHg)	150 ± 19
Diastolic Blood Pressure (mmHg)	81 ± 13
Pulse Pressure (mmHg)	69 ±17
Preoperative LV ejection fraction (%)	54 ± 11
Postoperative LV ejection fraction (%)	52 ± 9
Vascular diseases
Coronary diseases	24 (50 %)
Degenerative aortic valvular diseases	12 (25 %)
Aorta aneurism	8 (17 %)
Lower limb arteriopathy	3 (6 %)
Unknown	1 (2 %)
Surgical interventions
Coronary bypass	31 (65%)
Valve replacement	7 (14 %)
Lower limb vascular surgery	10 (21 %)

Data are mean ± SD or the percentage (%) and number (n) of patients.

Table 2 depicts the clinical characteristics of the 3 age- and gender-adjusted groups of patients for which blood samples were available. The 18 patients from the Surgery Group showed higher percentages of dyslipidemia and diabetes but no difference in prevalence of hypertension. Blood pressure levels were also similar in all 3 groups of subjects. No differences were observed between patients with asymptomatic atherosclerotic plaques and control subjects.

Table 2.

Clinical characteristics of 18 surgery patients with leukocyte TRF measurements as compared with age- and gender-matched groups of subjects with or without carotid plaques.

Description of population	Group 1 Surgery Group	Group 2 Atherosclerotic No surgery	Group 3 the Controls
Number of patients	18	18	18
Age (years)	67 ± 9	67 ± 9	67 ± 9
Gender
- Men	13	13	13
- Women	5	5	5
Medical history
Dyslipidemia	15 (83%)	6 (33%)	6 (33%)
Diabetes mellitus	6 (33%)	1 (6%)	0
Hypertension	11 (65%)	13 (72%)	13 (72%)
Current or ex-smoker	7 (41%)	7 (39%)	5 (30%)
Obesity	6 (35%)	10 (56%)	7 (39%)
Arterial parameters
Systolic Blood Pressure (mmHg)	151 ± 15	144 ± 19	148 ± 17
Diastolic Blood Pressure (mmHg)	79 ± 17	85 ± 11	85 ± 14
Pulse Pressure (mmHg)	72 ± 19	60 ± 18	64 ± 17
TRF length (kb)	6.15 ±0,98	6.21 ± 0,75	6.40 ± 0,5

Data are mean ± SD or the percentage (%) and number (n) of patients.

Analysis of correlation between Leukocyte telomere length (LTL) and the presence of atherosclerosis

Analysis of LTL in subjects with available blood samples revealed that, despite a trend toward shorter TRF in the group undergoing surgery (6.15 ± 0.98 kb) versus the control group (6.41 ± 0,5 kb), LTLs were not significantly different between the two groups (p = 0, 15). The group of patients with asymptomatic carotid atherosclerosis showed intermediate values (6.21 ± 0.75 kb). Upon re-distributing the population into 2 groups (atherosclerotic group, n = 36 and non atherosclerotic group, n = 18), LTLs measured in the atherosclerotic group were shorter (6.18 ± 0.13 kb) than those observed in the age- and sex-adjusted non-atherosclerotic group (6.41 ± 0.18 kb), although this difference was also not significant (p = 0.25).

In the control group, a negative correlation was observed between TRF and age (r = 0.64; p = 0.004). This same trend was also observed for LTL in patients with carotid atherosclerosis as well as in surgery patients, although the latter correlations were not significant (data not shown). For the entire population, the annual telomere attrition was estimated at approximately 44 bp /year.

Telomere length and aortic stiffness in vascular tissues

Tissue-specific analysis of telomere length showed that age-adjusted telomere length was longer in the saphenous vein (9.25 kb ± 0.28 (SEM) kb; n = 22) and the internal mammary artery (8.81 ± 0.24 kb; n = 29) than in the aortic axis (8.08 ± 0.27 kb; n = 23) (p < 0.013, aorta versus the other 2 segments) (figure 1a). There was no significant difference between saphenous vein and internal mammary artery telomere length.

Figure 1.

a - Comparison between age-adjusted TRF of the saphenous vein, internal mammary artery and aorta axis. Data are presented as the mean ± standard deviation. b - Comparison between age-adjusted TRF of healthy aortic axis versus pathologic aortic axis. Data are presented as the mean ± standard deviation

Upon dividing the aortic segments into those with macroscopic atherosclerotic lesions and those without lesions, telomere length was significantly shorter in aortic segments with atherosclerotic lesions (7.64 ± 0.32 kb; n = 11) versus those without (8.57 ± 0.30 kb; n = 13; p < 0.05) (figure 1b).

Discussion

To our knowledge, this is the first study measuring telomere length in leukocytes and in situ in both atherosclerotic and non atherosclerotic segments in humans. The main findings of this work were as follows: (1) arterial segments which do not develop atherosclerosis, such as the saphenous vein and internal mammary artery, have a longer telomere length than aortic segments; (2) telomere length is shorter in aortic tissues which present atherosclerotic lesions comparatively to corresponding tissues without atherosclerotic lesions.

Alterations associated with aging in blood vessels include an increase in vascular inflammatory response, in part due to increasing oxidative stress which promotes atherogenesis. Indeed, the implication of oxidative stress on replicative senescence and telomere attrition rate has been demonstrated in vitro (14). Moreover, we have recently shown similar results in an in vivo mouse model with short telomeres in which chronic glutathione inhibition increased oxidative stress and was able to induce a tissue-specific reduction in telomere length (17). Studies in cultured vascular smooth muscle cells (VSMCs) from atherosclerotic plaques have also shown local DNA damage due to increased oxidative stress (6), while senescence of cultured endothelial cells (ECs) isolated from the internal mammary arterial segments was accelerated in vitro by oxidative stress in atherosclerotic patients with high cardiovascular risks factors (18). The association between telomere attrition and the atherosclerotic process has also been evaluated in vitro in these studies whereby cultured senescent VSMCs and ECs were observed to exhibit markedly shorter telomeres (6, 18, 19); this shortening of telomere length was also reported in coronary endothelial cells from patients with coronary artery diseases (7). The above results led these authors to suggest that telomere shortening may participate in arterial ageing and its manifestations. In vivo, age-dependent telomere attrition rate has been studied in human abdominal aorta, and was found to be greater in both intima and media of the distal versus proximal aorta, while telomere length was negatively correlated with atherosclerotic grade (20). Altogether, these various studies allow us to conclude that oxidative stress exerts its effect on both atherogenic and non atherogenic vessels, and likely participates in the observed increase in telomere attrition. In the present study, telomere length was measured on all cells constituting the arterial wall. The shortened telomere length observed in atherogenic tissues (aortic axis) compared to those without atherosclerotic lesions may partly be explained by oxidative stress. Indeed, this shorter length is more pronounced in atherosclerotic plaques due to inflammation, and may likely accelerate telomere attrition of VSMCs in situ [6], similarly to other cells involved in the atherosclerotic process (ECs, inflammatory cells).

Our second major finding was that non atherogenic tissues, such as the saphenous vein and the internal mammary artery, exhibited longer telomeres compared to atherogenic tissues. Chang and Harley (21) have shown that due to hemodynamic stress, veins may present a slower telomere attrition rate than arteries. Our results show similar telomere length in an artery which, as in veins, does not develop atheroma. This is a key finding suggesting that influences other than hemodynamic factors are likely involved in explaining local TRF regulation in vascular beds. Comprehension of these mechanisms leading to both longer telomeres and the absence of atheroma could contribute to a better understanding of this specificity of certain arterial segments such as mammary or radial arteries.

Several limitations of our study should be noted. First, it was impossible to obtain arterial and blood tissues in all of the subjects. Thus, in order to be able to compare the various samples, all telomere measurements had to be adjusted for age and gender. The small number of studied samples may also explain the absence of a statistically significant difference in LTL between atherosclerotic patients and controls despite the presence of a clear trend to that effect.

In conclusion, the present data suggest an in situ regulation of telomere length in different arterial tissues with or without atheroma. These results also suggest tissue regulation of telomere size by local factors probably related to oxidative stress responses.

Acknowledgments: This study has been supported by a grant of the Institut de l'Atherothrombose, France. The work has also been supported by grants of INSERM (U961) and FRM (Fondation pour la Recherche Médicale, France; grant Number DCV20070409250). We are grateful to the staff of Department of Cardiovascular Surgery of the Nancy University hospital, France. We also thank Mr Pierre Pothier for his very pertinent comment on this paper and for linguistic corrections.

Funding sources: Supported by a grant of the „Institut de l'Atherothrombose“, and by the „Fondation pour la Recherche Médicale“ (FRM), France.

Conflict of Interest: None