Major depressive disorder (MDD) is associated with a significantly increased risk of developing serious medical illnesses that are more commonly seen with advanced age, such as diabetes, cardiovascular disease, immune impairments, stroke, dementia, osteoporosis, diabetes and metabolic syndrome 
and of dying significantly earlier (even after accounting for socio-demographic factors, suicide and risk factors such as smoking, alcohol and physical illness) 
. Indeed, major depression has been likened to a state of “accelerated aging,” with an increased incidence of aging-related illnesses 
. Various explanations for a prematurely aged phenotype in depression have been proposed, such as the “glucocorticoid cascade” hypothesis 
and “allostatic load” 
, as well as an unhealthy lifestyle or environment. In this study, we explored the additional (and not mutually exclusive) possibility that “accelerated aging” in depression occurs at the level of telomeres, as manifest in blood leukocytes. Further, we hypothesized that such changes are directly correlated with cumulative lifetime exposure to depression and are related to specific cytotoxic biochemical mediators, such as pro-inflammatory cytokines and oxidative stressors, which are often elevated in depression.
Telomeres are deoxyribonucleic acid (DNA)-protein complexes that cap the ends of the linear chromosomal DNA, protecting the genome from damage. In mitotic cells, telomeres can shorten with each division, unless this can be counteracted or reversed by the telomere-lengthening enzyme, telomerase 
. When telomeric DNA reaches a critically short length, as in cells undergoing repeated mitotic divisions (e.g., leukocytes 
and stem cells 
), cells become susceptible to senescence or apoptosis 
. Even in non-dividing cells, telomere shortening has been associated with cytotoxic stressors such as oxidative stress, which preferentially damages telomeric DNA compared with non-telomeric DNA, and chronic inflammation 
. Such enhanced telomere shortening also increases cellular susceptibly to apoptosis and death 
. Telomere length is emerging as a prognostic marker of disease risk and a robust indicator of human “biological age” (as distinguished from chronological age). It may represent a cumulative log of factors such as the number of cell divisions and of exposure to cytotoxic processes such as excessive oxidation and inflammation 
. Several recent studies in non-depressed populations have demonstrated an inverse relationship between leukocyte telomere length and the risk of current and future medical illnesses (such as cardiovascular and infectious diseases and dementia) and early death 
. Accelerated telomere shortening in individuals with MDD could help explain the increased medical morbidity seen in depression 
In addition to biochemical stressors such as oxidation and inflammation, chronic psychological stress is also associated with shortened telomeres 
. For example, healthy chronically stressed individuals (e.g., maternal caregivers of chronically ill children and family caregivers of demented individuals) showed significantly shorter leukocyte telomere length compared to age-matched controls 
, and in one study, telomere length was inversely correlated with the chronicity of caregiving (i.e., women with greater cumulative duration of caregiving stress had shorter telomeres) 
. This is consistent with the hypothesis that telomere shortening develops over time in proportion to cumulative exposure to the stressors. The difference in mean telomere base pairs (bp) between the two groups in that study suggested approximately 9–17 years of accelerated biological aging in the stressed, compared to the non-stressed, women 
. In the first study of telomere length in affective illnesses, including MDD, subjects also had significantly shortened leukocyte telomere length compared to controls, with an estimated acceleration of biological cell aging of over 10 years (average telomere shortening
660 bp) 
. That study utilized an extremely chronically ill population (average cumulative lifetime duration of illness
31.8±11.2 [SD] years). To the extent telomere length reflects cumulative exposure to depression and cytotoxic factors, telomere length should be inversely correlated with the lifetime chronicity of depression, and therefore, extrapolation of findings from that study to individuals with less chronic major depression may not be possible. A second study examined leukocyte telomere shortening in major depression and also found shortened telomeres 
, but it did not report the chronicity of depression in their sample. A very recent study also found shortened leukocyte telomeres in MDD (although various psychiatric, neurological and somatic disorders were not excluded in their sample), with an average shortening of 350 bp 
, representing approximately 6–8 years of “accelerated aging.” In that study, average telomere length was not related to lifetime depression history, as assessed by the time from the first onset of MDD through the time of the present assessment, but intervening periods of time that the subjects were not depressed were not excluded in that assessment. In the present study, to test effects of depression status and chronicity on cell aging, we evaluated depressed individuals across a broad range of depression chronicity and assessed chronicity as the estimated time actually spent in depressive episodes.
In addition to investigating telomere length in depression, we were interested in assessing biological factors associated with telomere shortening. Since depression is often associated with increased oxidative stress 
and with a pro-inflammatory milieu 
, we reasoned that oxidative stress and inflammation might contribute to shortened telomeres in depression, just as they are believed to do in certain medical illnesses and in preclinical models 
. We hypothesized that depressed individuals would have shorter leukocyte telomeres than matched controls, that telomere length would be inversely correlated with cumulative lifetime exposure to depression, and that telomere length would be inversely correlated with oxidative stress and inflammatory markers.
Subjects gave written informed consent to participate in this study, which was approved by the University of California, San Francisco (UCSF) Committee on Human Research (CHR).
To determine whether leukocyte telomeres are shortened in individuals with Major Depressive Disorder (MDD), whether this is a function of depression chronicity and whether this is related to putative mediators, oxidation and inflammation.
Eighteen subjects with MDD, diagnosed with the Structured Clinical Interview for DSM-IV-TR (SCID) 
, and 18 individually-matched healthy controls (matched by sex, ethnicity and age ±3 years) were recruited. One healthy control had an inadequate blood sample collection, leaving 18 depressed subjects and 17 healthy controls with usable data. Depressed subjects were all outpatients; they and the controls were recruited by fliers, bulletin board notices, Craigslist postings, newspaper ads and, in the case of depressed subjects, clinical referrals. Subjects were paid for their participation. SCID diagnostic interviews were conducted by an experienced clinical psychologist and were clinically verified by a separate psychiatric interview with a Board-certified psychiatrist. Depressed subjects with psychosis or bipolar histories were excluded, although co-morbid anxiety disorders were allowed when the depressive diagnosis was considered to be the primary diagnosis. Subjects with Post-Traumatic Stress Disorder (PTSD) were excluded, since PTSD may have important differences in stress hormone regulation 
. Seven of the depressed subjects had co-morbid anxiety diagnoses as follows: three with generalized anxiety disorder, two with obsessive-compulsive disorder, two with binge eating disorder (one of whom was in remission) and one with social anxiety disorder. Healthy controls were also screened with the SCID, and were required to have no present or past history of any DSM-IV Axis I or Axis II diagnosis. Potential subjects were excluded if they met SCID criteria for alcohol or substance abuse within 6 months of entering the study. Subjects in both groups were medically healthy (assessed by physical examination, review of systems and screening laboratory tests), had no acute illnesses or infections, and had not had any vaccinations within 6 weeks of entering the study. All subjects (depressed and control) were free of any psychotropic medications, including antidepressants, antipsychotics and mood stabilizers, as well as any hormone supplements, steroid-containing birth control or other interfering medications (e.g. statins) or Vitamin supplements above the U.S. Recommended Daily Allowances (e.g. Vitamin C, 90 mg/day), for a minimum of 6 weeks before entry into the study (with the exception of short-acting sedative-hypnotics, as needed, up to a maximum of 3 times per week, but none within one week prior to testing).
Subjects were admitted as outpatients to the UCSF Clinical and Translational Science Institute's Clinical Research Center at 8:00 am, having fasted (except water) since 10:00 pm the night before. Before proceeding with testing, all subjects were required to test negative on a urine toxicology screen (measuring the presence of abused drugs) and, in women of child-bearing capacity, a urine pregnancy test. After the subjects had sat quietly for 45 minutes, blood samples were obtained for leukocyte telomere length, oxidative stress markers (F2-isoprostanes and the anti-oxidant, Vitamin C) and inflammation (IL-6). Whole blood was drawn for the telomere length assay, and buffy coat was saved for leukocyte telomere length assay. DNA was prepared from whole blood using commercially available reagents (Gentra Puregene Blood Kit, Qiagen, Valencia, CA). Blood for the F2-isoprostane assay was collected into EDTA tubes with no vacuum, and blood for the Vitamin C assay was collected into foil wrapped serum separator tubes. Blood for IL-6 assay was collected into serum separator tubes.
Severity of depression in the depressed subjects was ascertained with the observer-rated 17-item Hamilton Depression Rating Scale (HDRS) 
. Total lifetime duration of depression was estimated in the depressed subjects using the life history methods of Sheline 
and Post 
, supplemented with information derived from the SCID interview and the Antidepressant Treatment History Form (ATHF) 
, which documents depressive episode durations as well as durations of antidepressant treatment, including the doses used and the treatment response. Only periods of time during which subjects were actively depressed (i.e., met DSM-IV criteria for MDD) were counted; periods of time during which subjects were not depressed were not included. Clinical history-taking and telomere assays were performed blind to each other.
High molecular weight DNA was extracted from frozen whole blood using commercially available reagents (Puregene, Gentra Systems, Qiagen, Valencia, CA). DNA quality and quantity were assessed with a nanodrop spectrophotometer and random samples were also assessed by agarose gel electrophoresis. The telomere length measurement assay was adapted from the published original method 
. Briefly, the T (telomeric) and S (single copy gene) values of each sample were determined by quantitative polymerase chain reaction (PCR) using the following primers: tel1b
[5′-CGGTTT(GTTTGG)5GTT-3′] and tel2b
[5′-GGCTTG(CCTTAC)5CCT-3′] for T and hbg1
[5′ GCTTCTGACACAACTGTGTTCACTAGC-3′] and hbg2
[5′-CACCAACTTCATCCACGTTCACC-3′] for S (human beta-globin). Genomic DNA from HeLa cells was used as the reference to quantify the T and S values relative to the reference DNA sample by the standard curve method. All PCRs were carried out on a Roche Lightcycler 480 real-time PCR machine with 384-tube capacity (Roche Diagnostics Corporation, Indianapolis, IN). The telomere thermal cycling profile consisted of: cycling for T (telomeric) PCR: denature at 96°C for 1 second, anneal/extend at 54°C for 60 seconds, with fluorescence data collection, 30 cycles; cycling for S (single copy gene) PCR: denature at 95°C for 15 seconds, anneal at 58°C for 1 second, extend at 72°C for 20 seconds, 8 cycles; followed by denature at 96°C for 1 second, anneal at 58°C for 1 second, extend at 72°C for 20 seconds, hold at 83°C for 5 seconds with data collection, 35 cycles. The inter-assay coefficient of variation (CV) for telomere length measurement was 4%. Details of the method can be found in 
The recommended approach for evaluating oxidative stress assesses the balance between anti-oxidants and oxidative by-products, with assessment of at least one antioxidant and one oxidized molecule 
. Ascorbic acid (Vitamin C) is often the preferred antioxidant to measure in this context 
. Although ascorbic acid is an essential nutrient in humans (entirely derived from dietary consumption), ascorbic acid levels reach steady state concentrations under fasting conditions (approximately six hours following food consumption), and they have been found to be relatively stable within individuals across time and to predict long-term health outcomes up to 12 years later 
. Because of this, and in order to limit the problem of multiple hypothesis testing by examining oxidative by-products and anti-oxidants separately, we a priori
defined “oxidative stress” as the ratio of F2-isoprostanes 
(a class of major oxidative by-products) to ascorbic acid 
. In a prior study from our group, the ratio of oxidative by-products to anti-oxidants was found to be inversely correlated with leukocyte telomere length in stressed caregivers 
F2-isoprostanes (a collection of isomers) were measured by Dr. Jason Morrow's Lab at Vanderbilt University by gas chromatography-mass spectrometry (GC-MS) as described previously. 
. The isoprostanes were extracted and purified with solid phase extraction and thin layer liquid chromatography and then converted to trimethylsilyl ether derivatives and analyzed by GC-MS. Ascorbic acid was measured by Kronos Science Laboratory using high performance liquid chromatagraphy (HPLC) method as described previously 
. Briefly, the serum sample was preserved by adding an equal volume of metaphosphoric acid and treated with dithiothreitol The resulting supernatant was injected into the HPLC system equipped with diode array detector (DAD). The separation was carried out using a 250×4.6 mm Capcell Pak NH2
column (Shiseido, Tokyo, Japan) with a flow rate of 1 mL/min of the mobile phases consisted of 10 mM ammonium acetate at pH 4.2 and methanol. Ascorbic acid was quantified using external standards with UV spectrophotometric detection at 243 nm wavelength.
Samples were collected in 10 ml serum separator tubes (SST) tubes (Becton Dickinson, Franklin Lakes, NJ). Serum was frozen and stored at – 80 C. A high sensitivity enzyme-linked immunosorbent assay was used to quantify IL-6 concentrations (R&D Systems, Minneapolis, MN). The assay sensitivity is <0.1 pg/ml, and average intra- and inter-assay coefficients of variation are 7% and 8% respectively. Each sample was analyzed in duplicate according to manufacturer protocol. IL-6 assays were performed in the lab of Dr. Firdaus Dhabhar at Stanford University.
We first assessed the impact of age, sex, body-mass index (BMI), and lifetime and current tobacco use as potential confounds; we found significant effects of age and sex on telomere length, and significant effects of age and BMI on IL-6 (data presented below). Lifetime and current tobacco use were not significantly related to any of these variables. Consequently, all analyses were controlled for age and sex, and analyses involving IL-6 were additionally controlled for BMI. Before analyzing the data, distributions were examined for normality; non-normal distributions were natural log transformed (Ln).
Between-group comparison of the demographic variables was by independent sample t-tests, Chi square tests and independent sample Kruskal-Wallis tests. Independent sample tests, rather than paired sample tests, were used, since the absence of blood data in one control subject (described above) resulted in unequal group sizes. Other between-group data, when covariates were applied, were analyzed by analysis of covariance (ANCOVA). Correlations between variables were assessed by linear regression, or by hierarchical linear regression, with age and sex (and BMI, in the case of IL-6) entered first. All tests were 2-tailed with an alpha