Rapid telomere erosion in schizophrenia

Telomeres are DNA repeats located at the ends of chromosomes, which progressively shorten with each cell division, thus serving as a molecular biomarker of physical aging in many organisms.1 Unexpectedly, telomere dysfunction was recently described in psychological stress2 and mood disorders.3 These intriguing findings suggest a molecular link between neural activity and aging, and raise the possibility that telomere maintenance may be impaired in other psychiatric disorders.

Here, we tested the hypothesis that telomere dynamics are altered in schizophrenia, a severe psychotic disorder. Subjects with schizophrenia were recruited from two cohorts, A and B, and were evaluated using structured interviews (Supplementary Methods). Telomere lengths were measured in a total of 51 individuals with schizophrenia, 52 unaffected, unrelated individuals, and 24 unaffected family members. Genomic DNA was isolated from peripheral blood lymphocytes and telomere length was measured using quantitative PCR.4 Significance was evaluated using analysis of variance (ANOVA) or analysis of covariance (ANCOVA) where necessary (Supplementary Methods). Since telomere lengths were quantitated in two independent laboratories, the results were analyzed separately, as described below.

We first examined telomere lengths in 31 individuals diagnosed with schizophrenia (DSM-IV), recruited from the two separate cohorts (A, n=13; B, n=18), 41 unrelated individuals screened as having no major psychiatric disorder (Control) and 24 unaffected family members (Family), most of whom were first-degree relatives of cohort B (Figure 1a; Supplementary Table S1A). As expected, there was a negative correlation of telomere length with age (r=−0.28, P=0.01). Since age and sex can affect telomere length,5 the data were analyzed using an ANCOVA, controlling for the effects of these two variables. We observed a significant main effect of diagnosis (F=6.44, df=2, P=0.002), but there was no main effect of age or sex within this analysis. No significant difference in telomere lengths was observed between the two schizophrenia cohorts, although they were evaluated using different structured diagnostic interviews. In post hoc analyses, individuals with schizophrenia had significantly shorter mean telomere lengths than controls (P=0.002) (Figure 1a; Supplementary Table S1A).

Figure 1

Reduced telomere length in peripheral blood lymphocytes of individuals with schizophrenia. A measure of telomere length, T/S,4 was determined for individuals in various diagnostic groups. (a) Comparison of unrelated, unaffected individuals (Control, n=41), unaffected individuals related to individuals with schizophrenia in cohort B (Family, n=24) and individuals diagnosed with schizophrenia combined from two cohorts, A and B (Sz, n=31). T/S was transformed to natural logarithms to permit comparisons of normally distributed values within each diagnostic group (Supplementary Table S1A). *P=0.002 for Control vs Sz using ln(T/S). (b) T/S was determined for unrelated, unaffected individuals (Control, n=26) and individuals with schizophrenia (Sz, n=33) in an age-matched data set containing only male subjects (Supplementary Table S1B). #P=0.008 for Control vs Sz. Means±s.e.m. are depicted.

In a second study, we analyzed a group in which age was matched and all individuals were of a single sex, male. Our rationale was that telomere attrition is not linear at all ages and can differ between the sexes.5 Thus, we recruited 20 additional male individuals with schizophrenia and added them to cohort A (total n=33). In addition, we recruited 12 unaffected males and added them to the existing male controls (total n=26). Telomere lengths were measured and analyzed separately from the previous study. Again, we observed that individuals with schizophrenia had significantly shorter mean telomere lengths than unaffected controls (P=0.008) (Figure 1b; Supplementary Table S1B).

Chronic antipsychotic treatment could affect telomere length. However, there was no correlation between current antipsychotic dose (r=−0.071, P=0.66) and estimated lifetime antipsychotic dose (r=0.045, P=0.78) with telomere length (Supplementary Figure S1).

Telomere erosion appears to be occurring rapidly in individuals with schizophrenia. By their mid- to late 30's, the mean T/S (a measure of telomere length4), of individuals with schizophrenia was lower than controls by about 0.37–0.41 (Figures 1a and b; Supplementary Tables S1A and B), which corresponds to approximately 1580–1750 bp telomeres (see Supplementary Methods). Since telomere lengths were measured after about 20 years of illness in schizophrenia (mean age 36.2–38.8 years), we estimate that the rate of telomere loss has more than doubled with the onset of schizophrenia, and that lymphocytes in individuals with schizophrenia had aged the equivalent of at least 25 years more than their control counterparts. It should be noted that these estimates are based on cross-sectional studies, and longitudinal measurements carried out on the same individuals over time would be required to obtain a more accurate rate of telomere erosion.

The presence of significant telomere loss in schizophrenia implies an abnormal aging component in this disorder. Although the cause is unknown, rapid telomere loss can be caused by stress,2 oxidative damage5 and DNA instability,1, 5 mechanisms that have been implicated in schizophrenia pathology.6, 7 These findings therefore suggest new directions of research that may illuminate the neuropathology of this enigmatic disorder.


  1. 1

    Blackburn EH . FEBS Lett 2005; 579: 859–862.

    CAS  Article  Google Scholar 

  2. 2

    Epel ES, Blackburn EH, Lin J, Dhabhar FS, Adler NE, Morrow JD et al. Proc Natl Acad Sci USA 2004; 101: 17312–17315.

    CAS  Article  Google Scholar 

  3. 3

    Simon NM, Smoller JW, McNamara KL, Maser RS, Zalta AK, Pollack MH et al. Biol Psychiatry 2006; 60: 432–435.

    CAS  Article  Google Scholar 

  4. 4

    Cawthon RM . Nucleic Acids Res 2002; 30: e47.

    Article  Google Scholar 

  5. 5

    von Zglinicki T, Martin-Ruiz CM . Curr Mol Med 2005; 5: 197–203.

    CAS  Article  Google Scholar 

  6. 6

    Yao JK, Reddy RD, van Kammen DP . CNS Drugs 2001; 15: 287–310.

    CAS  Article  Google Scholar 

  7. 7

    Malaspina D, Harlap S, Fennig S, Heiman D, Nahon D, Feldman D et al. Arch Gen Psychiatry 2001; 58: 361–367.

    CAS  Article  Google Scholar 

Download references


This work was supported by the National Institutes of Health (R01 MH070898 to BP; R01 NS047209 to H-TK; R21 MH071720 to LED; R01 MH044292 to Jurg Ott), the Stanley Research Medical Foundation (BP) and NARSAD.

Author information



Corresponding author

Correspondence to H-T Kao.

Additional information

Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kao, H., Cawthon, R., DeLisi, L. et al. Rapid telomere erosion in schizophrenia. Mol Psychiatry 13, 118–119 (2008). https://doi.org/10.1038/sj.mp.4002105

Download citation

Further reading