Obesity, cigarette smoking, and telomere length in women

Abstract

Obesity and smoking are important risk factors for many age-related diseases. Both are states of heightened oxidative stress, which increases the rate of telomere erosion per replication, and inflammation, which enhances white blood cell turnover. Together, these processes might accelerate telomere erosion with age. We therefore tested the hypothesis that increased body mass and smoking are associated with shortened telomere length in white blood cells. We investigated 1122 white women aged 18-76 years and found that telomere length decreased steadily with age at a mean rate of 27 bp per year. Telomeres of obese women were 240 bp shorter than those of lean women (p=0.026). A dose-dependent relation with smoking was recorded (p=0.017), and each pack-year smoked was equivalent to an additional 5 bp of telomere length lost (18%) compared with the rate in the overall cohort. Our results emphasise the pro-ageing effects of obesity and cigarette smoking.

Full text

Research Letters
Telomeres cap the ends of chromosomes and protect
them from degradation and end-to-end fusion.
Telomeres of cultured somatic cells undergo erosion
with each cycle of replication, and oxidative stress
enhances this process.1
Both obesity and cigarette smoking are important risk
factors in many age-related diseases, and are associated
with increased oxidative stress and inflammation.2,3 The
latter process is marked by increased white blood cell
(WBC) turnover. Telomere attrition (expressed in
WBCs) can serve as a marker of the cumulative oxidative
stress and inflammation and, consequently, show the
pace of biological ageing. We therefore expected obese
individuals and smokers to have shortened telomeres.
To investigate this hypothesis we studied WBC telomere
length in 1122 healthy white women aged 18–72 years,
examining the relations with both smoking and obesity-
related phenotypes.
Participants were female twins (45 monozygotic and
516 dizygotic pairs) from the TwinsUK Adult Twin
Registry, a group previously developed to study the
heritability and genetics of diseases with a higher
prevalence among women. These women were recruited
from the general population through national media
campaigns in the UK, and were similar to age-matched
population singletons in terms of disease-related and
lifestyle characteristics (www.twinsuk.ac.uk).4In our
cohort, body-mass index (BMI) was 30 in 119 (11%)
women and 20 in 85 (8%). None of the participants
had clinical diabetes. 531 (47%) women had never
smoked, 369 (33%) were ex-smokers, 203 (18%) were
still smoking, and smoking status was unknown for 19
(2%). Smoking history was recorded with a standardised
questionnaire. Smoking exposure was measured as
pack-years=number of cigarette packs smoked per
daynumber of years smoking.
A venous blood sample was taken after an overnight
fast. We extracted DNA from WBCs, and measured
the concentration of leptin in serum with a
radioimmunoassay (Linco, St Charles, MO, USA). We
measured the mean of the terminal telomere restriction
fragment (TRF) lengths, an index of telomere length,
with the Southern blot method.5Written and oral
informed consent was obtained from all participants.
The St Thomas’ Hospital Research Ethics Committee
approved the study.
Standard linear regression techniques were used to
correlate the TRF length with age and the age-adjusted
TRF with individual factors. Log-transformed leptin
values were used for both the age-adjusted and
unadjusted linear regressions. The associations between
categorical variables and telomere length, adjusting for
age or other covariates, were assessed using analyses of
variance. To adjust for non-independence between twins
in a pair, bootstrap sets were generated by selecting a
random twin from each pair using analysis of variance,
and the p value of the mean test statistic from
100 replicates was used to confirm statistical
significance. S-Plus 6.0 (Insightful Corp) software was
used. No significant difference in the variables studied
was noted between monozygotic and dizygotic twins.
Telomere length decreased steadily with age at a mean
rate of 27 bp per year (SD 50·2; figure A) and a highly
significant negative correlation was detected (table). The
proportion of the variance in telomere length accounted
for by age was 20·6%. Squared and cubed age terms
were also added to the model and had no significant
effect on telomere length (p=0·92 and p=0·98,
respectively) suggesting a linear relation between TRF
and age.
Additionally, we noted no clear age-related difference
in rates of TRF loss; average rate of loss was 27·7 bp per
year in women aged 50 years and over and 25·7 bp
per year in those younger than 50 years.
BMI, leptin concentration in serum, and smoking
status were all significantly correlated with age (r= 0·12,
r=0·13, r=–0·10, respectively). Leptin concentration in
serum and BMI were strongly positively correlated
(r=0·76). However, the correlations between smoking
status and BMI (r=–0·05) and between leptin
concentration in serum and smoking status (r=–0·06)
were not statistically significant. BMI, leptin
concentration in serum, and packs-year of cigarettes
smoked were negatively correlated with telomere length.
The regression coefficients of these variables remained
statistically significant after adjustment for age (table).
Published online
June 14, 2005
DOI:10.1016/S0140-6736(05)
66630-5
Twin Research and Genetic
Epidemiology Unit, St Thomas’
Hospital, London, UK
(A M Valdes PhD, T Andrew PhD,
E Oelsner BSc, L F Cherkas PhD,
Prof T D Spector MD); and
Hypertension Research Center,
University of Medicine and
Dentistry of New Jersey,
Newark, New Jersey, USA
(J P Gardner PhD, M Kimura MD,
Prof A Aviv MD)
Correspondence to:
Professor T D Spector
tim.spector@kcl.ac.uk
www.thelancet.com Published online June 14, 2005 DOI:10.1016/S0140-6736(05)66630-5 1
Obesity, cigarette smoking, and telomere length in women
A M Valdes, T Andrew, J P Gardner, M Kimura, E Oelsner, L F Cherkas, A Aviv, T D Spector
Obesity and smoking are important risk factors for many age-related diseases. Both are states of heightened oxidative
stress, which increases the rate of telomere erosion per replication, and inflammation, which enhances white blood
cell turnover. Together, these processes might accelerate telomere erosion with age. We therefore tested the
hypothesis that increased body mass and smoking are associated with shortened telomere length in white blood
cells. We investigated 1122 white women aged 18–76 years and found that telomere length decreased steadily with
age at a mean rate of 27 bp per year. Telomeres of obese women were 240 bp shorter than those of lean women
(p=0·026). A dose-dependent relation with smoking was recorded (p=0·017), and each pack-year smoked was
equivalent to an additional 5 bp of telomere length lost (18%) compared with the rate in the overall cohort. Our
results emphasise the pro-ageing effects of obesity and cigarette smoking.

Research Letters
In addition to the linear models tested on continuous
measures, lean individuals were found to have
significantly longer telomeres than women with mid-
range BMIs, who, in turn, had longer telomeres than
obese individuals (figure, B; p=0·026).
Age-adjusted telomere length was negatively
correlated with log-transformed leptin concentration in
serum (table) and the mean age-adjusted telomere
length showed a progressive decrease through the
quartiles of leptin concentration (figure, C).
Individuals who had never smoked had longer age-
adjusted telomeres than former smokers and both had
longer telomeres than current smokers (figure, D;
p=0·02). Moreover, age-adjusted telomere length
decreased with the amount of cigarettes smoked (table;
figure, D). Each pack-year smoked was equivalent to a
loss of an additional 5 bp, or 18% of the average annual
loss in age-adjusted telomere length, compared with the
rate in the overall cohort.
No statistical interaction between leptin concentration
and smoking history or between BMI and smoking
history was noted. After fitting stepwise linear
regression, age, smoking (p0·0004), and leptin
(p0·006) remained significantly associated with
telomere length, but BMI did not, suggesting that the
mechanisms by which obesity affects telomere length
might be better represented by leptin concentration than
by BMI. We conclude that both obesity and smoking are
associated with shortened WBC telomere length in
women. Additionally, telomere length was inversely
correlated with the serum concentration of leptin—
a marker and regulator of body fat that itself may have
some pro-inflammatory properties known to increase
oxidative stress.6
Our findings suggest that obesity and cigarette
smoking accelerate human ageing. Our cross-sectional
data underscore the considerable variation in telomere
length between individuals. Thus large cohorts are
needed to capture the effects of inflammation and
oxidative stress.1However, in view of the hypothesis that
telomere length in vivo represents cellular turnover and
exposure to oxidative and inflammatory damage, the
difference in telomere length between being lean and
being obese corresponds to 8·8 years of ageing; smoking
(previous or current) corresponds on average to
4·6 years of ageing; and smoking a pack per day for
40 years corresponds to 7·4 years of ageing. Our results
emphasise the potential wide-ranging effects of the two
most important preventable exposures in developed
countries—cigarettes and obesity.
Contributors
A M Valdes participated in the statistical analysis and in the preparation
of the manuscript. T Andrew participated in the processing and
statistical analysis of the data. E Oelsner and L F Cherkas collected and
verified the clinical information of the study participants. J Gardner and
M Kimura did the telomere assays and participated in the processing of
the data. T D Spector and A Aviv designed and coordinated the study
and participated in the preparation of the manuscript.
2www.thelancet.com Published online June 14, 2005 DOI:10.1016/S0140-6736(05)66630-5
Lean (BMI 20)
(n=85)
BMI 20–30
(n=918)
Obese (BMI 30)
(n=119)
BMI category
TRF length (kb)
Q1 Q2 Q3 Q4
Quartiles of serum leptin
TRF length (kb)
Non-smoker
(n=531)
Ex-smoker
(n=369)
Current smoker
(n=203)
Smoking category
TRF length (kb)
6·8
7·2
7·0
7·4
7·6
7·8
6·8
7·2
7·0
7·4
7·6
7·8
6·8
7·2
7·0
7·4
7·6
7·8
6·8
7·2
7·0
7·4
7·6
7·8
Q1 Q2 Q3 Q4
Quartiles of cigarette pack-years
TRF length (kb)
5·0
5·5
6·0
6·5
7·0
7·5
8·0
8·5
9·0
9·5
10 20 30 40 50 60 70 80
Age (years)
TRF length (kb)
A
BC
DE
8·0 8·0
8·0 8·0
Figure: Relation between telomere length and (A) age, (B) BMI, (C) leptin, (D) smoking history, and
(E) cigarette pack-years
Data for B–E are age-adjusted mean TRF with SD.
Mean (SD) Correlation p Age-adjusted correlation p*
with TRF with TRF
TRF (kb) 7·06 (0·67)
Age (years) 47·77 (12·11) –0·455 0·0001 n/a n/a
BMI (kg/m2) 25·05 (4·69) –0·126 0·0001 –0·077 0·031
Serum leptin (ng/mL) 16·26 (12·50) –0·124 0·0001 –0·088 0·019
Smoking status† n/a –0·031 ns –0·087 0·017
Cigarette pack-years‡ 8·15 (14·31) –0·214 0·0001 –0·110 0·045
ns=not significant. *Statistical significance of regression coefficient from 100 bootstrap replicates. †Coded as 0=never smoked,
1=ex-smokers, 2=current smokers. ‡Among ex-smokers and current smokers only.
Table 1: Descriptive statistics of study subjects and correlations with telomere terminal restriction
fragment (TRF) length before and after adjusting for chronological age

Research Letters
Conflict of interest statement
We declare that we have no conflict of interest.
Acknowledgments
We thank all the twins participating in TwinsUK, R Swaminathan of
Chemical Pathology for the leptin assays, and Gabriela L Surdulescu for
the DNA preparation. This work was supported by the Wellcome Trust
Functional Genomics Initiative for the TwinsUK program and WT
project grant no 07495/t/Z/04/Z, the Healthcare Foundation of New
Jersey and NIH grant AG021593. The sponsor of the study had no role
in study design, data collection, data analysis, data interpretation, or
writing of the report. The corresponding author had full access to all the
data in the study and had final responsibility for the decision to submit
for publication.
References
1 Aviv A. Telomeres and human aging: facts and fibs. DOI:
10·1126/sageke.2004·51.pe43 (accessed May 25, 2005).
2 Burke A, Fitzgerald GA. Oxidative stress and smoking-induced
vascular injury. Prog Cardiovasc Dis 2003; 46: 79–90.
3 Dandona P, Alijada A, Bandyopandhyay A. Inflammation: the link
between insulin resistance, obesity and diabetes. Trends Immunol
2004; 25: 4–7.
4 Andrew T, Hart DJ, Snieder H, de Lange M, Spector TD,
MacGregor AJ. Are twins and singletons comparable? A study of
disease-related and lifestyle characteristics in adult women.
Twin Res 2001; 4: 464–77.
5 Benetos A, Okuda K, Lajemi M, et al. Telomere length as an
indicator of biological ageing: the gender effect and relation with
pulse pressure and pulse wave velocity. Hypertension 2001; 37:
381–85.
6 Beltowski J, Wojcicka G, Jamroz A. Leptin decreases plasma
paraxonases 1 (PON1) activity and induces oxidative stress: the
possible novel mechanism for proatherogenic effect of chronic
hyperleptinemia. Atherosclerosis 2003; 170: 21–29.
www.thelancet.com Published online June 14, 2005 DOI:10.1016/S0140-6736(05)66630-5 3