Telomere attrition in sickle cell anemia

Telomeres are specialized nucleoprotein structures at the end of chromosomes, involved in genome stability and cellular functioning.1 Telomere(definition) shortening leads to cellular senescence(definition) and is considered as a biomarker of aging.1 Telomere attrition and the resulting cellular senescence could play a role in pathologic states associated with chronic inflammation or oxidative stress, as seen during sickle cell disease (SCD).2, 3 SCD associates hemolytic and viscosity-related complications, leading to decreased life expectancy.4 The single gene mutation responsible for abnormal sickle hemoglobin (HbS) contrasts with the phenotypic heterogeneity observed among SCD patients. The objectives of this study were to evaluate whether telomere attrition is associated with SCD, its phenotypic presentations, and outcomes. Between 2003 and 2013, consecutive homozygous SCD patients over 15 years of age were enrolled with one or more of their family member with an AA genotype (controls). The study was approved by institutional review board and all patients signed written informed consent. Clinical and laboratory data were measured during a routine visit at steady state, ie >2 months from a painful crisis, and >3 months from transfusion or hydroxyurea intake. Follow-up information for survival (up to 10 years) was obtained by medical chart review and interview of the family and referring physician. All SCD patients and control subjects underwent blood sampling for genotype determination and biological assessment—blood cell count, hemoglobin electrophoresis, and red blood cell (RBC) density, assessed by measuring the percentage of dense RBCs (density >1.120) and the median density of RBCs. The telomere length was measured in white blood cells by using quantitative PCR.5 The telomere repeat copy number to single-gene copy number (T/S) ratio was determined with a 7900HT thermocycler (Applied Biosystems, Foster City, CA) in a 384-well format, using the comparative Ct method (T/S = 2–ΔΔCt). Continuous data were expressed as mean ± standard deviation and were compared using the Student T-test or Mann–Whitney test. Categorical variables, expressed as percentages, were evaluated using the chi-square test or Fisher exact test. To evaluate the association between telomere length and age, we used the Pearson r correlation coefficient. SCD patients were further separated into two groups with longer or shorter telomere length, based on the mean value of the T/S ratio in this group. Survival curves were generated using the Kaplan–Meier method and compared using the log-rank test. Mortality over the follow-up period was also analyzed using multivariable Cox model including factors yielding a P-value <.10 by univariate analysis. A total of 446 subjects were analyzed, including 318 SCD homozygous patients and 128 family member controls with AA genotype. Overall, the telomere length of SCD patients was lower than that of controls (1.25 ± 0.50 vs. 2.10 ± 0.65, P < .0001), despite a comparable mean age (31 ± 9 vs. 35 ± 14 years, P = .1). Figure 1A depicts the relation between telomere length and age by decades in SS and AA individuals. There was a negative correlation between telomere length and age in controls (r = −0.34; P < .001) but not in SCD patients (r = −0.04; P = .5). The effect of age on telomere length was significantly influenced by genotype (P < .001 for interaction). SCD patients with shorter telomeres had more frequent iron therapy, lower values of HbS percentage and dense RBCs, higher values of total hemoglobin concentration and oxygen saturation, and more pain crisis during the preceding year than their counterparts with longer telomeres (Supporting Information Appendix). A total of 20 SCD patients (6%) died during follow-up, which lasted 8.7 ± 1.9 years. Long-term survival was reduced in SCD patients with shorter telomeres as compared to those with longer telomeres (P = .03 for log rank test, Figure 1B), and this effect persisted in univariate and multivariate Cox regression analysis (hazard ratio [95% CI] of 3.01 [1.10-8.27], P = .03, and 4.35 [1.24-15.29], P = .02, respectively). Box and Whiskers plots of telomere repeat copy number to single-gene copy number (T/S) ratio measured in adult sickle cell disease patients (SS, gray bars) and controls (AA, white bars) per decades of life. B, Kaplan–Meier analysis of the probability of survival of sickle cell disease patients with longer or shorter telomere length Our finding of shorter telomeres in SCD patients as compared to controls is in accordance with one previous report6 but contradicts another study reporting longer telomeres in SCD patients as compared to healthy controls.7 The inclusion in the latter study of patients treated with hydroxyurea (known to affect telomere maintenance) and those having received recent blood transfusion may in part explain this discrepancy. Telomere length is determined genetically, vary with lifespan, and can be altered by inflammation and oxidative stress. Younger individuals with SCD showed the more striking difference between observed and expected telomere length (as anticipated by values in the control group), a feature also seen in individuals carrying mutations in telomerase genes. Despite the common familial origin of cases and controls, a pure genetic effect associated with the SS mutation may at least in part explain shorter telomeres in SCD patients. The telomere length regulator gene TNKS was identified by genome wide association studies as a genetic modifier of the severity of SCD.8 Enhanced inflammatory and pro-oxidant activity in SCD patients3 might also favor telomere attrition.2 The profile of SCD patients with shorter telomeres in our study suggests a hyperviscosity/vaso-occlusive phenotype rather than a chronic hemolytic one4: these patients had higher values of hemoglobin, lower values of HbS, and were more prone to pain crises in the year preceding blood sampling. The exact mechanism through which telomere a