ORIGINAL CONTRIBUTIONS

Weight Loss Associated with Bariatric Surgery Does Not RestoreShort Telomere Length of Severe Obese Patients After 1 Year

Caterina Formichi & Silvia Cantara & Cristina Ciuoli & Ornella Neri &Francesco Chiofalo & Federico Selmi & Andrea Tirone & Giuseppina Colasanto &

Leonardo Di Cosmo & Giuseppe Vuolo & Furio Pacini

# Springer Science+Business Media New York 2014

AbstractBackground Telomere shortening is physiologically associat-ed with ageing but it may be influenced by oxidative stressand chronic inflammation, linked to obesity. Thus, obesitymight represent an additional cause of telomere attrition. Weaim to study relative telomere length (RTL) in obese subjects

with and without metabolic syndrome and to assess the effectof weight loss induced by bariatric surgery.Methods We evaluated RTL in 107 obese subjects (62 withmetabolic syndrome and 45 without metabolic syndrome),compared to 130 age-matched non-obese subjects. We alsomeasured RTL in a subgroup of 93 obese patients prior to and3, 6, 9 and 12 months after surgery.Results RTL of obese subjects was significantly shorter(p<0.0001) than non-obese subjects but without differencesbetween patients with and without metabolic syndrome(p=0.19). RTL was significantly shorter than baseline at3, 6, 9 and 12 months after bariatric surgery.Conclusions These results confirm that obese subjects haveshorter telomeres compared to non-obese subjects, but RTL isnot influenced by the presence of metabolic syndrome. RTLshows an additional attrition during the immediate post-operative period, probably due to a catabolic state.

Keywords Telomere length . Obesity . Bariatric surgery .

Metabolic syndrome

Introduction

Telomeres are DNA structures composed of tandem repeats ofTTAGGG sequence and associated proteins, at the ends ofhuman chromosomes. The presence of telomeres avoids end-to-end fusions during cellular division and prevents ends ofchromosome from being recognized as double-strand break byDNA repair proteins [1]. Telomeres progressively shortens ateach cellular division, contributing to cellular senescence andapoptosis. Telomere length, thus, represents a dynamic regula-tor of cellular life span and a marker of cellular ageing [2].Telomere length shows a high inter-individual variability, partlydetermined by genetic factors and partly due to environmentalfactors; among the latter, oxidative stress and inflammation,

C. Formichi : S. Cantara : C. Ciuoli (*) :O. Neri : F. Chiofalo :F. Selmi : F. PaciniDepartment of Medical, Surgical and Neurological Sciences,University of Siena, Viale Mario Bracci 16, 53100 Siena, Italye-mail: c.ciuoli@ao-siena.toscana.it

C. Formichie-mail: catefo@libero.it

S. Cantarae-mail: silviacantara@hotmail.com

O. Nerie-mail: orni@hotmail.it

F. Chiofaloe-mail: fra_chiofalo@libero.it

F. Selmie-mail: f.selmi@live.it

F. Pacinie-mail: pacini8@unisi.it

A. Tirone :G. Colasanto : L. Di Cosmo :G. VuoloBariatric Surgery Unit, Department of Surgical Sciences,University of Siena, Viale Mario Bracci 16, 53100 Siena, Italy

A. Tironee-mail: andrea.tirone@alice.it

G. Colasantoe-mail: mykene@libero.it

L. Di Cosmoe-mail: dicosmo@unisi.it

G. Vuoloe-mail: g.vuolo@ao-siena.toscana.it

OBES SURGDOI 10.1007/s11695-014-1300-4

which represent the major contributors to ageing and ageing-related diseases, are known to accelerate telomere attrition [3].

Obesity is a chronic disease, characterized by a state ofincreased oxidative stress and inflammation [4, 5]. Therefore,oxidative stress and inflammation have been suggested as anunderlying mechanism for the association between obesity andshort telomeres [6]. Several large studies have shown thatshorter telomeres are associated with obesity in adulthood[6–9] and with several anthropometric measures of obesity,such as body mass index (BMI) and waist and hip circumfer-ence [6, 7, 10, 11]. Most of the associations identified to datehave been in adult pre-menopausal women. In contrast, otherstudies of comparable size found no associations betweenleukocyte telomere length and BMI, adiposity measures oradipokine levels [12–14]. Chronic inflammation is also recog-nized as the central pathophysiological mechanism involved inthe genesis of metabolic diseases [4]. Therefore, it seems rea-sonable that the presence of metabolic syndrome could deter-mine an additional telomere attrition. Many recent studies haveassociated shorter telomeres with several metabolic disorders,such as type 2 diabetes and hypercholesterolemia [15–19].

Data on the effects of weight loss on telomere length in obesesubjects are scarce and conflicting. Healthy lifestyles and dietsseem to be positively correlated with telomere length [1]. In astudy evaluating telomere length in subcutaneous adipose tissuecells, formerly obese patients showed shorter telomeres thannever-obese subjects, suggesting that this is an established andirreversible feature of obesity that could contribute to its comor-bidities [20]. Similarly, in a recent study on post-menopausalwomen, telomere length did not change after 12 months ofdietary weight loss and physical exercise [21]. By contrast,O’Callaghan and colleagues demonstrated that telomere lengthincreased up to tenfold in a cohort of obese men after dietaryweight loss and gain of telomere length appeared to be greater ifmore weight and body fat was lost [22]. Chronic dietary restric-tion, assessed by self-report, may even be a risk factor forpremature telomere shortening [23], and frequent weightcycling has been associated with shorter telomere length [8].

Clinical evidence suggests that only bariatric surgery hasachieved substantial long-term weight loss in morbid obesity[24]. Some evidences suggest that oxidative stress markersand inflammatory markers decrease after bariatric surgery[25, 26]. Thus, we aimed to evaluate the effect of meta-bolic syndrome and rapid weight loss, induced by bariatricsurgery, on telomere length.

Materials and Methods

Patients

We evaluated 107 obese patients (22 male and 85 female; meanBMI ± DS 46.47±8.5, range 33.3–78.7 kg/m2; mean age±DS

42.43±10.9 years, range 19–66), followed in the Section ofEndocrinology of University of Siena (Italy), and 130 non-obese subjects (74 male and 56 female; mean BMI ± DS23.4±2.7, range 20.1–24.8 kg/m2; mean age±DS 43.85±13.7 years, range 16–79), free from metabolic and autoim-mune diseases. The two groups were homogeneous for age.According to ATP III criteria [27], we subdivided the cohortof obese subjects in 62 patients with metabolic syndrome(mean BMI ± DS 47.93±8.6) and 45 patients without meta-bolic syndrome (mean BMI ± DS 44.98±8.1). As expected,all the features of metabolic syndrome were significantlydifferent between the two groups (Table 1). The proportionof diabetic subjects was 45.2 % in the group of obesesubjects affected by metabolic syndrome and 6.7 % in thegroup of obese subjects without metabolic syndrome.

Out of 107 obese patients, 93 underwent bariatric surgery(39 sleeve gastrectomies, 22 gastric bandings, 25 gastric by-passes, 5 biliopancreatic diversions, 2 gastric plications); ofthese patients, 75 were available for evaluation at 3 monthsfrom surgery, 79 at 6 months, 31 at 9 months and 62 at12 months after surgery.

DNA Extraction

Blood samples were obtained from patients after informedconsent in accordancewith local ethical committee guidelines.Genomic DNAwas extracted from peripheral blood leukocytesusing QIAamp DNAMini Kit (Qiagen). DNA concentrationwas assessed in a fluorometric manner (Glomax Multi JR,Promega). For each sample, a stock solution of 10 ng/20 μlwas prepared.

Table 1 Anthropometric measures and metabolic data in obese patientswith and without metabolic syndrome

MS (n=62) No MS (n=45) p

Age (years) 45.8±8.8 37.6±11.6 0.0001a

BMI (kg/m2) 47.9±8.6 44.9±8.1 0.049a

Waist circumference (cm) 125.9±15.5 115±12.2 0.009a

HDL Cholesterol (mg/dl) 42.7±10.6 55.2±12.9 <0.0001a

Triglycerides (mg/dl) 189.9±113 88.9±26.9 <0.0001a

Glicemia (mg/dl) 125.2±54.5 93.0±12.4 <0.0001a

Insulin (mIU/l) 24.7±29.4 14.1±8 0.0038a

HOMA-IR 6.9±8.5 3.1±2.7 0.0021a

Hypertension 27/60 (45 %) 3/41 (7 %) <0.0001b

Smokers 29/58 (50 %) 22/42 (52 %) 0.841

MS metabolic syndrome, BMI body mass index, HDL high-densitylipoprotein, HOMA-IR homeostasis model of assessment insulinresistancea Student’s t-testb Fisher’s exact test

OBES SURG

Measurements of Relative Telomere Length

This analysis was performed using quantitative PCR (Q-PCR)technique. Q-PCR assay was performed as previously de-scribed [28, 29] on 50 ng/μl of genomic DNA. Telomerelength quantification involved determining the relative ratioof telomere (T) repeat copy number to a single-copy gene (S)copy number (T/S ratio) in experimental samples using stan-dard curves. This ratio is proportional to the average telomerelength. 36B4, encoding acidic ribosomal phosphoprotein P0,has been used as the single-copy gene. Primers have been asfollows and were used at 300-nM final concentration in areaction mix containing 12.5 μl SYBER Green PCR MasterMix for a final volume of 25 μl: telomere sense, 5′-GGTTTTTTGAGGGTGAGGGTGAGGGTGAGGGTGAGGGT-3′;telomere antisense, 5′-TCCCCGACTATCCCTATCCCTATCCCTATCCCTATCCCTA-3′; 36B4 sense, 5′-CCCATTCTATCACAACGGTACAA-3′; and 36B4 antisense, 5′-CAGCAAGTGGGAAGGTGTAATCC-3′. A standard curve was in-cluded in each run. The thermal cycling profile for the telo-mere amplification was 95° for 10 min followed by 30 cyclesof 95 °C for 15 s and 54 °C for 1 min, and for the 36B4amplification, it was 95° for 10 min followed by 40 cycles of95 °C for 15 s and 60 °C for 1 min. To exclude the presence ofnon-specific binding between SYBER Green and primers, amelting curve was added at the end of all PCR amplificationreactions. The melting profile was as follows: 3 min at 95 °Cfollowed 15 s at 95 °C (40 repetitions), 30 s at 60 °C, 30 s at55 °C and 10 s at 55 °C.

Statistics

Statistical analysis was performed with GraphPad Prismavers.5.00.288. One-way ANOVA with Bonferroni’s post-testwas used to determine the differences in telomere lengthamong groups. For the analysis of pre- and post-surgicalrelative telomere length (RTL), for each patient and at eachtime point, the paired t test was used. To compare variablesamong categories, Fisher’s exact test was used. Correlationtests used was linear regression analysis. All data are presentedas mean±SD. A p value <0.05 was considered significant.

Results

Relative telomere length, expressed as T/S ratio, was signifi-cantly shorter (p<0.0001) in obese patients (mean±SD 0.94±0.28) compared to normal-weight subjects (mean±SD 1.87±0.95) (Fig. 1).

As expected, in our group of healthy subjects, RTL de-creased with ageing (p=0.0003) (Fig. 2a) but we did not findthis correlation in the obese group (Fig. 2b).

According to ATP III criteria, we obtained a group of 62patients with metabolic syndrome and a group of 45 patientswithout metabolic syndrome. Mean RTL of patients withmetabolic syndrome (mean±SD 0.97±0.29) was not signifi-cantly different from mean RTL of patients without metabolicsyndrome (mean±SD 0.91±0.27) (Fig. 3).

Furthermore, in the whole cohort of obese subjects,telomere length did not correlate with any of the metabolicparameter studied, such as BMI, hypercholesterolemia,glycaemia and insulin resistance, expressed as HOMA index(data not shown).

In the group of 93 patients submitted to bariatric surgery,RTL was significantly shorter at 3, 6, 9 and 12 months aftersurgery compared to basal pre-surgical levels (mean±SD0.88±0.18, p=0.01; mean±SD 0.84±0.16, p=0.0005;mean±SD 0.81±0.15, p<0.0001; mean±SD 0.79±0.14,p<0.0001, respectively) (Fig. 4). We did not find any signifi-cant correlation between the percent change in RTL and EWL,percent change in BMI and waist circumference both at 6and 12 months after surgery. Results were not differentwhen we performed linear regression analysis using a logof RTL distribution.

Discussion

Our results confirmed previous report that telomere length isshorter in obese patients compared to normal-weight subjects.In addition, telomere length correlated with age, as expected,in control group but not in the obese group. One possibleexplanation for this finding is that the burden of oxidativestress and inflammation linked to obesity is too strong to beaffected by chronological age. It is also possible that, assuggested by Kim et al., duration of obesity may be moreimportant than weight change per se [8]. Alternatively, thismolecular feature might represent one of the many genetic

Fig. 1 Relative telomere length in obese (closed square) and non-obese(closed circle) subjects. Mean±SD is presented

OBES SURG

signatures associated with obesity. Assuming that adiposetissue accumulation increases oxidative stress and causes de-regulation of inflammatory cytokines, we decided to studywhether a significant weight loss could improve telomerelength in a subgroup of obese patients followed up 1 year afterbariatric surgery. Surprisingly, we found that RTL declined at3, 6, 9 and 12 months from surgery. The additional attritionobserved in the post-operative period is probably due to thesevere post-surgical catabolic state. Catabolic state is knownto increase the burden of oxidative stress and accelerate telo-mere erosion [30]. To confirm this hypothesis, we measuredIGF1 levels, which is known to be an anabolic hormone [30],and we found a significant decrease of IGF1 levels in theimmediate post-surgical levels (p<0.01) compared to baselineand a recovery to baseline levels after 12 months (data notshown). However, we did not find any significant correlationbetween IGF1 levels and the amount of RTL decrease at eachtime point neither between IGF1 levels and RTL value at eachtime point. Whether RTL will return to the level of normal-weight subjects on long-term period should be evaluated instudies with longer follow-up. However, the lack of improve-ment of telomere length after surgery-induced weight loss is

consistent with previous reports evaluating the effect of die-tary weight loss on telomere length [20, 21].

We hypothesized that the short follow-up may negativelyinfluence RTL because in such a short period, several param-eters known to influence telomere length are not stabilized yet,particularly weight loss. In addition, other factors may affectRTL independently from follow-up. Telomere length is acomplex trait that is shaped by a combination of genetic,epigenetic and environmental determinants, such as psycho-logical stress, low physical activity levels, smoking and so-cioeconomic status; however, the range of factors that influ-ence telomere dynamics is not fully established [9]. Amongenvironmental determinants of telomere length, lifestyle mayalso play a key role; it has been shown that dietary restraint,especially in combination with a strong tendency to overeat, isrelated to perceived stress and to shorter leukocytes telomeres,independent of body mass index [23], probably due to expo-sure to both psychological and metabolic stress, including animbalance between anabolic and catabolic hormones [30].

Another finding of our study was the lack of associationbetween shorter RTL and the presence of metabolic syndromein the obese group. An association could have been expected

Fig. 2 a Linear regressionbetween RTL and age (expressedin years) in non-obese subjects.b Linear regression between RTLand age (expressed in years) inobese patients

Fig. 3 Relative telomere length in obese patients with metabolic syn-drome (closed circle) and obese patients without metabolic syndrome(closed square). Mean±SD is presented

Fig. 4 Relative telomere length in obese patients submitted tobariatric surgery at 0, 3, 6, 9 and 12 months after surgery. Mean±SDis presented

OBES SURG

based on the assumption that metabolic syndrome represents astrong inflammatory status, known to affect telomere length.The lack of this association in our study might be due toselection of patient with severe obesity (BMI>35). It is pos-sible that in this category of severe obesity, even in theabsence of metabolic syndrome, the inflammatory status isenough to influence telomere length. To this purpose, it will beuseful to study the RTL in patients with different degrees ofobesity.

Some potential limitations of our study are the lack ofinformation on markers of oxidative stress or inflammation,so we could not demonstrate the presence of a relationshipwith RTL, and, as discussed above, the short follow-up whichcould affect the results because somemetabolic parameters arenot yet stabilized 1 year after surgery.

In conclusion, the short telomere length associated withsevere obesity is not influenced by the presence of metabolicsyndrome and is not ameliorated by weight loss induced bybariatric surgery, at least in the short-term outcome.

Conflict of Interest The authors have nothing to declare.

References

1. Paul L. Diet, nutrition and telomere lenght. J Nutr Biochem.2011;22(10):895–901.

2. Barceló A, Piérola J, López-Escribano H, et al. Telomere shorteningin sleep apnea syndrome. Respir Med. 2010;104(8):1225–9.

3. Tzanetakou IP, Katsilambros NL, Benetos A, et al. “Is obesity linkedto aging?”: adipose tissue and the role of telomeres. Ageing Res Rev.2012;11(2):220–9.

4. Furukawa S, Fujita T, Shimabukuro M, et al. Increased oxidativestress in obesity and its impact on metabolic syndrome. J Clin Invest.2004;114(12):1752–61.

5. Lumeng CN, Saltiel AR. Inflammatory links between obesity andmetabolic disease. J Clin Invest. 2011;121(6):2111–7.

6. Valdes AM, Andrew T, Gardner JP, et al. Obesity, cigarette smoking,and telomere length in women. Lancet. 2005;366(9486):662–4.

7. Nordfjall K, Eliasson M, Stegmayr B, et al. Telomere length isassociated with obesity parameters but with a gender difference.Obesity (Silver Spring). 2008;16(12):2682–9.

8. Kim S, Parks CG, DeRoo LA, et al. Obesity and weight gain inadulthood and telomere length. Cancer Epidemiol Biomarkers Prev.2009;18(3):816–20.

9. Cassidy A, De Vivo I, Liu Y, et al. Associations between diet,lifestyle factors, and telomere length in women. Am J Clin Nutr.2010;91(5):1273–80.

10. Lee M, Martin H, Firpo MA, et al. Inverse association betweenadiposity and telomere length: the fels longitudinal study. Am JHum Biol. 2011;23(1):100–6.

11. Gardner JP, Li S, Srinivasan SR, et al. Rise in insulin resistance isassociated with escalated telomere attrition. Circulation. 2005;111(17):2171–7.

12. Bekaert S, De Meyer T, Rietzschel ER, et al. Telomere length andcardiovascular risk factors in a middle-aged population free of overtcardiovascular disease. Aging Cell. 2007;6(5):639–47.

13. Bischoff C, Petersen HC, Graakjaer J, et al. No association betweentelomere length and survival among the elderly and oldest old.Epidemiology. 2006;17(2):190–4.

14. Diaz VA, Mainous AG, Player MS, et al. Telomere length andadiposity in a racially diverse sample. Int J Obes (Lond).2010;34(2):261–5.

15. Harte AL, da Silva NF, Miller MA, et al. Telomere length attrition, amarker of biological senescence, is inversely correlated with triglyc-erides and cholesterol in south Asian males with type 2 dibatesmellitus. Exp Diabetes Res. 2012;2012:895185.

16. Salpea KD, Humphries SE. Telomeres length in atherosclerosis anddiabetes. Atherosclerosis. 2010;209(1):35–8.

17. Saliques S, Zeller M, Lorin J, et al. Telomere length and cardiovas-cular disease. Arch Cardiovasc Dis. 2010;103(8–9):454–9.

18. Fitzpatrick AL, Kronmal RA, Gardner JP, et al. Leukocyte telomerelength and cardiovascular disease in the cardiovascular health study.Am J Epidemiol. 2007;165(1):14–21.

19. Demissie S, Levy D, Benjamin EJ, et al. Insulin resistance, oxidativestress, hypertension, and leukocyte telomere length in men from theFramingham Heart Study. Aging Cell. 2006;5(4):325–30.

20. Moreno-Navarrete JM, Ortega F, Sabater M, et al. Telomere length ofsubcutaneous adipose tissue cells is shorter in obese and formerlyobese subjects. Int J Obes (Lond). 2010;34(8):1345–8.

21. Mason C, Risques RA, Xiao L, Duggan CR, Imayama I,Campbell KL, Kong A, Foster-Schubert KE, Wang CY, AlfanoCM, Blackburn GL, Rabinovitch PS, McTiernan A. Independentand combined effects of dietary weight loss and exercise onleukocyte telomere length in postmenopausal women. Obesity(Silver Spring). 2013.

22. O’Callaghan NJ, Clifton PM, Noakes M, et al. Weight loss inobese men is associated with increased telomere length anddecreased abasic sites in rectal mucosa. Rejuvenation Res.2009;12(3):169–76.

23. Kiefer A, Lin J, Blackburn E, et al. Dietary restraint and telomerelength in pre- and postmenopausal women. Psychosom Med.2008;70(8):845–9.

24. Picot J, Jones J, Colquitt JL, et al. The clinical effectiveness andcosteffectiveness of bariatric (weight loss) surgery for obesity: asystematic review and economic evaluation. Health TechnolAssess. 2009;13(41):1–190. 215–357.

25. Rao SR. Inflammatory markers and bariatric surgery: a meta-analysis. Inflamm Res. 2012;61(8):789–807.

26. Dadalt C, Fagundes RL, Moreira EA, et al. Oxidative stress markersin adults 2 years after Roux-en-Y gastric bypass. Eur J GastroenterolHepatol. 2012;25(5):580–6.

27. Expert Panel on Detection, Evaluation, and Treatment of High BloodCholesterol in Adults. Executive Summary of the Third Report ofthe National Cholesterol Education Program (NCEP) ExpertPanel on Detection, Evaluation, and Treatment of High BloodCholesterol in Adults (Adult Treatment Panel III). JAMA.2001;285(19):2486–97.

28. Cawthon RM. Telomere measurement by quantitative PCR. NucleicAcids Res. 2002;30(10):e47.

29. Capezzone M, Cantara S, Marchisotta S, et al. Short telomeres,telomerase reverse transcriptase gene amplification, and increasedtelomerase activity in the blood of familial papillary thyroid cancerpatients. J Clin Endocrinol Metab. 2008;93(10):3950–7.

30. Epel ES. Psychological and metabolic stress: a recipe for acceleratedcellular aging? Hormones (Athens). 2009;8(1):7–22.

OBES SURG