Telomere length, cardiovascular risk and arteriosclerosis in human kidneys: an observational cohort study

Population characteristics

Cohort 1 consisted of 217 unique kidney donors with sufficient amount of good-quality leucocytic DNA available for evaluation of telomere length. In this cohort, 144 pre-implantation kidney biopsies were available for histological evaluation. Fourteen biopsies were of insufficient quality according to the Banff 1997 criteria, leaving 130 baseline biopsies for histological analysis. On average, 24.0 ± 16.7 glomeruli were obtained per biopsy (range, 10–89).

Cohort 2, for evaluation of intrarenal telomere length, consisted of 40 kidney donors. Of these 40 subjects, good-quality DNA from leucocytes was available for 32 subjects and good-quality DNA derived from biopsies was available for all 40 kidneys. All 40 biopsies included in this cohort were of sufficient quality according to the Banff 1997 criteria for histological evaluation.

Table 1 summarizes the characteristics of these two cohorts and the histology of the biopsies that were included. Kidney function, expressed as eGFR, was 89.7 ± 40.9 (range 42.3–189.3) mL/min/1.73m² in Cohort 1 and 114 ± 42.0 (range 60.8–197.2) mL/min/1.73m2 in Cohort 2, respectively.

Calendar age, clinical demographics and leucocyte telomere length

Mean log T/S ratio of telomere length was 0.05 ± 0.25 (range −0.95 to 0.71). Leucocyte telomere length inversely correlated with older calendar age (r=−0.2, p = 0.005) (Table 2).

History of hypertension vs. no hypertension and history of cardiovascular events vs. no history also associated with shorter telomere length. Women had longer telomere length. Other clinical demographics, including history of diabetes mellitus, history of smoking, living versus deceased donation, brain death vs. cardiac death, ischemic and hemorrhagic stroke as reason for donor death, body mass index, and eGFR were not associated with leucocyte telomere length. In multivariate linear regression analysis, older calendar age, history of cardiovascular events, male gender and history of hypertension were independent explanatory factors for shorter telomere length (Table 2).

Leucocyte telomere length, clinical demographics and renal histology

In Cohort 1, arteriosclerosis in renal biopsies significantly associated with shorter leucocyte telomere length (log T/S ratio −0.3 ± 0.4 vs. 0.1 ± 0.2 in subjects with vs. without renal arteriosclerosis; p = 0.0008 (Figure 1B and Table 3). For one standard deviation increase of the log T/S ratio, independent of calendar age, gender and history of cardiovascular event, the odds of renal arteriosclerosis decreased by 64% (Odds ratio 0.36; 95% CI 0.17–0.77; p = 0.02). Moreover, arteriosclerosis in renal biopsies significantly associated with Δ biological age - calendar age (r = 0.3; p = 0.005).

Interstitial fibrosis, tubular atrophy, glomerulosclerosis and arteriolar hyalinosis did not associate with leucocyte telomere length. These histological lesions associated significantly with calendar age: for each 10 years increase in calendar age, the odds for interstitial fibrosis, tubular atrophy and glomerulosclerosis increased by 79%, 208% and 63%, respectively (Table 3).

To further describe the complex relations between calendar age, leucocyte telomere length and renal histology, principal component analysis was performed. The summery plot is presented in Figure 2A. This analysis illustrated a dichotomy in the correlation between the histological findings and telomere length versus calendar age. Glomerulosclerosis, tubular atrophy, interstitial fibrosis and arteriolar hyalinosis clustered with calendar age, while arteriosclerosis clustered with leucocyte telomere length.

Intrarenal telomere length, clinical demographics and renal histology

In Cohort 2, mean T/S ratio of intrarenal telomere length was 0.07 ± 0.09 (range −0.16–0.24). Intrarenal telomere length correlated significantly with leucocyte telomere length (r = 0.4, p = 0.001). Shorter intrarenal telomere length associated with history of hypertension (log T/S ratio 0.04 ± 0.02 vs. 0.10 ± 0.02 with vs. without hypertension; p = 0.05), but not with other demographics. Multivariate linear regression analysis was not performed in this smaller cohort to avoid overfitting of the data.

In this cohort, shorter intrarenal telomere length associated significantly with the presence of arteriosclerosis (log T/S ratio −0.04 ± 0.06 vs. 0.08 ± 0.01 with vs. without arteriosclerosis, p = 0.007) (Figure 1B). Intrarenal telomere length did not associate with any other histological parameter.

Principal component analysis was applied to further describe the complex relation between calendar age, renal histology, leucocyte telomere length and renal telomere length. The summery plots are presented in Figure 2B. This independent analysis confirmed the dichotomy in the correlation between the histological findings and telomere length versus calendar age, and the significant association between intrarenal telomere length and presence of renal arteriosclerosis.

Inclusion and exclusion criteria

All prospective deceased donors of a single kidney transplant in adult recipients, which were performed between March 2004 and December 2010 at the University Hospitals Leuven (Leuven, Belgium), were eligible for Cohort 1 of this study. Since March 2004, kidney grafts undergo routine biopsies prior to implantation (N = 352). In Cohort 2, donor kidneys transplanted after November 2012 were biopsied prior to implantation, and were used for evaluation of intrarenal telomere length (N = 40). This study was approved by the Ethics Committee/Institutional Review Board of the University Hospitals Leuven, Leuven, Belgium (OG032; ML7499 and ML9785; clinicaltrials.gov NCT01331668).

Clinical data collection

Clinical data were obtained from the Eurotransplant database (“Eurotransplant Donor Report”), which is maintained prospectively and is the central source of donor data for organ transplantation in the Eurotransplant region (www.donordata.eu). The following data were collected: calendar age, gender, cause of death, weight and length, living vs. deceased donor, brain death vs. cardiac death donor, body mass index, history of hypertension, diabetes mellitus, smoking, history of cardiovascular events prior to donation (including reason for death in deceased donors) and terminal serum creatinine levels before organ recovery. Renal function was estimated by the 4-variable Modification of Diet in Renal Disease (MDRD) equation (estimated glomerular filtration rate; eGFR) [19].

Kidney biopsies and histologic evaluation

One pathologist (EL) reviewed all pre-implantation kidney biopsies, without knowledge of any demographic information of the kidney donor. The biopsy specimens were wedge biopsies with slides containing 4 to 10 paraffin sections (2 μm) that were stained with hematoxylin eosin, periodic acid–Schiff, and a silver methenamine staining method (Jones). The severity of histologic lesions, interstitial fibrosis (Banf “ci” score), tubular atrophy (Banf “ct” score), arteriolar hyalinosis (Banf “ah” score) and arteriosclerosis (Banf “cv” score = vascular fibrous intimal thickening = intimal hyperplasia), were scored semiquantitatively according to the Banff criteria [20]. In addition, the total number of glomeruli in each biopsy, and the number of globally sclerosed glomeruli, were calculated separately. Only biopsies with > 10 glomeruli (A quality) were included for evaluation of glomerulosclerosis.

Telomere length in peripheral blood leucocytes

Peripheral blood samples for DNA extraction from leucocytes were routinely obtained from the donors at time of organ recovery. DNA was extracted using the simple salting out procedure for extracting DNA from human nucleated cells as described by S.A. Miller.21 Both DNA yield (ng/μL) and purity ratios A260/280 and A260/230 were determined and needed to be within strict quality limits (yield 50 ng/μL; purity ratio range 1.5–2 and 1.5–2 for A260/280 and A260/230, respectively) for inclusion of the samples in the study. Extracted DNA samples were stored at −80°C until further use. Leucocyte telomere length was measured as the telomere repeat copy number relative to two single gene nuclear control genes (36B4 and ACTB), by a modified version of a previously described PCR based assay [20, 22]. The forward and reverse primers were 5′-ACACTAAGGTTTGGGTTTGGGTTTGGGTTTG GGTTAGTGT-3′ and 5′-TGTTAGGTATCCCTATC CCTATCCCTATCCCTATCCCTAACA-3′, respectively. The forward and reverse primers for the reference genes were respectively 5′-ACTCTTCCAGCCTTCCT TCC-3′ and 5′-GGCAGGACTTAGCTTCCACA-3′ for ACTB, and 5′-GGAATGTGGGCTTTGTGTTC-3′ and 5′-CCCAATTGTCCCCTTACCTT-3′ for 36B4. The telomere reaction contained Fast SYBR^® Green I dye 2x (Applied Biosystems, Lennik, BE) mastermix, forward (100 nM) and reverse (900 nM) primer and 12.5 ng DNA. The telomere reactions were performed in triplicate. The thermal cycling profile for the telomere reaction consisted of following steps: 20 sec at 95°C, 2 cycles of 15 sec at 94°C and 15 sec at 49°C and 40 cycles of 15 sec at 94°C, 10 sec at 62°C and 15 sec at 74°C. For the reference genes a 10 μl PCR reaction contained Fast SYBR^® Green I dye 2x (Applied Biosystems, Lennik, BE) mastermix, forward (10 μM) and reverse (10 μM) primer and 12.5 μg DNA. Reactions were run in duplicate. Amplification specificity and absence of primer dimers was confirmed by melting curve analysis at the end of each run.

Each PCR-plate contained two inter-run calibrators and two no-template controls. For the telomere assay, reference sample were included, one with relative short telomeres and one with relative long telomere. After thermal cycling, raw data were collected and processed. Cq-values of telomere were normalized relative to the two reference genes using the qBase software (Biogazelle, Zwijnaarde, BE). The program uses modified software from the classic comparative CT method (ΔΔ CT) that takes into account multiple reference genes and uses inter-run calibration algorithms to correct for run-to-run differences [23]. Coefficient of variation for telomeres was 2.6%, for reference genes 1.6% and for the T/S ratio less than 7%. The corresponding numbers for tissue samples were less than 1% for telomeres and reference genes and 5% for the T/S ratio. Although this assay provides a relative measurement of telomere length, T/S ratios correlate well with absolute telomere lengths determined by Southern blot (r= 0.89, N= 20) [24]. Of the 352 consecutive kidney donors that were eligible for Cohort 1, 217 peripheral blood samples were included; one hundred thirty-five samples were not evaluated, either due to insufficient amount of leucocytic DNA available, or due to insufficient DNA quality for reliable telomere assessment. Accelerated biological aging was evaluated by calculating delta (Δ) biological age - calendar age. Δ (range −3 – 3) was calculated using the difference of quartile telomere length (0 = longer telomere length, 3 = shortest telomere length) - quartile calendar age (0 = youngest age group, 3 = oldest age group).

Telomere length in kidney biopsies

Since November 2012, next to the wedge biopsy that was used for histological evaluation, a full renal cortical biopsy core was obtained prior to implantation, and immediately stored in Allprotect Tissue Reagent (Qiagen, Venlo, The Netherlands) solution, until extraction. DNA extraction was performed by the Allprep DNA/RNA/miRNA Universal Kit (Qiagen, Venlo, The Netherlands) on a QIAcube instrument (Qiagen, Venlo, The Netherlands). Telomere length in renal tissue samples was measured by a modified version of the protocol for the peripheral blood samples. Telomere length was measured as the telomere repeat copy number relative to a single gene nuclear control genes (36B4). The telomere reaction mixture contained 1x Qiagen Quantitect Sybr Green Mastermix 2.5 mM of dithiothreitol, 300 nM of telomere forward primer (5′-ACACTAAGGTTTGGGTTTGGGTTTGGGTTTGGG TTAGTGT-3′), and 900 nM of telomere reverse primer (5′-TGTTAGGTATCCCTATCCCTATCCCTATCCC TATCCCTAACA-3′). The PCR ran for 1 cycle at 95°C for 10 min for activation of the DNA polymerase, followed by 2 cycles of 15 sec at 94°C and 2 min at 49°C, and 30 cycles of 15 sec at 94°C, 20 sec at 62°C and 1 min 40 sec at 74°C. All 40 biopsies included in Cohort 2 passed quality control for assessment of intrarenal telomere length.

Statistical analysis

Differences between groups were analyzed using parametric or non-parametric one-way ANOVA. Associations between the clinical demographics, telomere length and renal histology were assessed by means of linear or logistic regression analysis, as appropriate, as well as with principal component analysis. From the principal component analysis, a two-dimensional scatter plot was generated to represent the histologic lesions according to their score in the loading matrix. Multiple linear and binomial logistic regressions, with backward parameter selection, were used to model the determinants of telomere length. For backward parameter selection, the following variables were considered for entry into the model for the determinants of telomere length: calendar age, gender, history of hypertension, history of diabetes mellitus and body mass index, history of cardiovascular events, living vs. deceased donor, brain-death vs. cardiac death, ischemic stroke as reason for death, hemorrhagic stroke as reason for death and renal function (eGFR (mL/min/1.73m2). The associations between calendar age, telomere length and the different histological lesions were adjusted for gender and history of cardiovascular events in the multiple logistic regression analyses. All tests were two-sided and p values of less than 0.05 were considered to indicate statistical significance. The results are expressed as numerical values and percentages for categorical variables and as mean ± standard deviation for continuous variables, unless otherwise specified. Analyses were done with SAS (version 9.2; SAS institute, Cary, NC), JMP9.0 (SAS institute, Cary, NC) and GraphPad Prism (version 5.00; GraphPad Software, San Diego, CA) software.