Telomere Length Differently Associated to Obesity and Hyperandrogenism in Women With Polycystic Ovary Syndrome

Abstract

Background: Polycystic Ovary Syndrome (PCOS) often present metabolic disorders and hyperandrogenism (HA), facts that may influence the telomere length (TL).

Aims: To compare the absolute TL (aTL) between women with PCOS and control women, and their association with the presence of obesity and HA parameters.

Materials and methods: The PCOS group included 170 unrelated women outpatients and the control group, 64 unrelated donor women. Anthropometric, biochemical-clinical parameters and androgen profile were determined. The PCOS patients were divided accordingly to the presence of obesity and androgenic condition. The aTL was determined from peripheral blood leukocytes by Real Time quantitative PCR.

Results: Women with PCOS exhibited a significantly longer aTL than controls after age adjustment (p=0.001). A stepwise multivariate linear regression in PCOS women, showed that WC (waist circumference) contributed negatively (b=-0.17) while testosterone levels contributed positively (b=7.24) to aTL. The non-Obese PCOS (noOB-PCOS) presented the longest aTL when compared to controls (p=0.001). Meanwhile, the aTL was significantly higher in the hyperandrogenic PCOS phenotype (HA-PCOS) than in the controls (p=0.001) and non hyperandrogenic PCOS phenotype (NHA-PCOS) (p=0.04). Interestingly, when considering obesity and HA parameters in PCOS, HA exerts the major effect over the aTL as non-obese HA exhibited the lengthiest aTL (23.9 ± 13.13 Kbp). Conversely, the obese NHA patients showed the shortest aTL (16.5 ± 10.59 Kbp).

Conclusions: Whilst a shorter aTL could be related to the presence of obesity, a longer aTL would be associated with HA phenotype. These findings suggest a balance between the effect produced by the different metabolic and hormonal components, in PCOS women.

Keywords: hyperandrogenism; metabolic and endocrine disorders; obesity; polycystic ovary syndrome; telomere length.

Figure 1

Comparison of Absolute telomere length between control group and PCOS. (A) Distribution of absolute telomere length according to age in control and PCOS groups. (—): The linear regression for control group (R^{2 =} 0.15; Regression coefficient r=-0.28, p=0.25) and (- - -): PCOS group (R^{2 =} 0.18; Regression coefficient r=-0.46; p=0.02). (B) Bar graphs illustrates mean ± standard error of mean (SEM) of aTL in control and PCOS groups. Univariate analysis adjusted by age (ANCOVA); p<0.05 was considered as significant. aTL, absolute telomere length; Kbp, kilobase pairs; PCOS, polycystic ovary syndrome.

Figure 2

Comparison of absolute telomere length between control group and PCOS according to the presence or absence of obesity (A) and hyperandrogenic condition (B). Bar graphs illustrates mean ± standard error of mean (SEM) of absolute telomere length for each group. noOB-PCOS, non-obese PCOS; OB-PCOS, obese PCOS women; NHA-PCOS, non hyperandrogenic PCOS; HA-PCOS, hyperandrogenic PCOS. Univariate analysis adjusted by age (ANCOVA), followed by Bonferroni post-hoc test for multiple factors or groups. NS, non significant difference; p<0.05 was considered as significant; aTL, absolute telomere length; Kbp, kilobase pairs.

Figure 3

Comparison of absolute telomere length between control and PCOS according to obesity and androgenic condition simultaneously presence. Bar graphs illustrates mean ± standard error of mean (SEM) of absolute telomere length for each group. noOB-NHA, non-obese non hyperandrogenic PCOS; noOB-HA, non-obese hyperandrogenic PCOS; OB-NHA, obese non hyperandrogenic PCOS; OB-HA, obese hyperandrogenic PCOS. Univariate analysis adjusted by age (ANCOVA), followed by Bonferroni post-hoc test applied for multiple factors or groups. aTL, absolute telomere length; p<0.05 was considered as significant; Kbp, kilobase pairs.