Telomere Length in Patients with Transfusion-Dependent Thalassemia

Background: Thalassemia is a hereditary hemolytic anemia with a severity ranging from mild, non transfusion-dependent to severe chronic anemia requiring lifelong transfusion. Transfusional iron overload is a major complication in patients with transfusion-dependent thalassemia (TDT). Telomeres are sequences of nucleotides forming the end caps of chromosomes that act as a DNA repair system. Iron overload in thalassemia can cause increased oxidative stress which leads to cellular damage and senescence. This may result in telomere length shortening. The degree of telomere length shortening may re�ect the severity of thalassemia. Methods: This research aimed to study the telomere length in patients with TDT in comparison to non-thalassemic individuals and to identify the clinical and laboratory parameters that are associated with telomere length. We conducted a cross-sectional study in patients with TDT aged ³18 years. Telomere length was measured by real-time quantitative PCR. Results: Sixty-ve patients with TDT were enrolled onto the study. There were 37 female patients (54.4%). The median age was 27 (18-57) years, and mean pre-transfusion hemoglobin (Hb) was 7.1 (± 1.07) g/dL. The mean telomeric terminal restriction fragment length (TRFL) of patients with TDT and the controls was 6.11 (± 0.61) kb and 6.79 (± 0.84) kb, respectively (p <0.0001). There was a signi�cant correlation between TRFL and age (p =0.0002), and Hb (p=0.044). There was no correlation of telomere length with other factors. Conclusions: Our study showed that TDT patients had shorter telomere length compared with controls. Telomere shortening in TDT was an aging-dependent process and associated with lower hemoglobin level.


Background
Thalassemia is a common hereditary hemolytic anemia in Thailand.About 30-40% of Thai population carries at least one of the abnormal globin genes [1].Clinical severity of thalassemia varies greatly and could be classi ed as non-transfusion dependent thalassemia (NTDT) which requires occasional blood transfusion in speci c situations and transfusion dependent thalassemia patients (TDT) which requires regular blood transfusion [2,3].Long-term blood transfusion and increased iron absorption from the gastrointestinal tract lead to iron overload, which can result in peroxidative damages to cells, organelles and membranes.The cellular damage associated with iron overload is mainly mediated by the state of oxidative stress and the effect of free oxygen radicals on various cell components and tissue [4][5][6].
Telomeres are the terminal section of chromosomes consisting of hexameric protein repeats amino acid (5'-TTAGGG-3').The telomeres play roles in stabilization of chromosomes, protection of DNA from end to end fusion and reparation of DNA.At each cell division, telomeres become shorter and nally reach a critically short length when their function will be lost.This means that cells can no longer divide and results in cellular aging and apoptosis [7,8].
The length of the telomeres in leukocytes can be used as an indication of good health, aging and longevity of the cell.Some human disorders associated with shorter telomere length originate from defective function of telomerase, an enzyme that extends the telomeres of chromosomes.Certain diseases originating from mutations in genes controlling the DNA repair system also result in accelerated telomere shortening and premature aging.One such condition is aplastic anemia [7][8][9][10].
There is limited data about telomere length in patients with thalassemia.Chaichompoo et al reported that the rate of telomere shortening in patients with beta-thalassemia/hemoglobin E disease was accelerated when compared to normal individuals.Oxidative stress is one of the factors contributing to erosion of telomeres and cell replication capability.Telomere shortening is associated with the clinical severity of thalassemia [11].
We conducted the study in patients with TDT who had severe clinical symptoms.Both alpha and beta thalassemia patients were included to de ne the clinical factors that were associated with telomere shortening.

Study population and de nition
We conducted a cross-sectional study in patients with TDT, aged 18 years and up, who attended Adult Hematology clinic at Chiang Mai University Hospital.TDT was de ned as a patient with thalassemia who required red cell transfusion at least 3 times per year.

Clinical and laboratory measurement
We collected clinical and laboratory parameters at the time of enrollment to identify factors associated with telomere length including age, sex, history of splenectomy, iron chelation, and factors associated with clinical severity including: pre-transfusion Hb level (mean steady-state Hb level in the previous 10 visits), units of red blood cell transfusion per month, serum ferritin (maximum, minimum, mean of serum ferritin at multiple times every year).All patients were investigated for telomere length, liver function test (LFT), non-transferrin bound iron (NTBI), hemolysis parameters, MRI for cardiac T2 star (T2*) and liver iron concentration (LIC).Thirty healthy individuals, age and sex-matched, were included as controls and were tested for telomere length.

Telomere length measurement
Telomere length was measured by using real-time quantitative PCR.DNA was extracted from peripheral blood leukocytes and the telomere repeat sequence copy number to single-copy gene copy number was compared.The speci city of all ampli cations was determined by melting curve analysis.
The results of telomere length from PCR are expressed as T/S (telomere/single copy gene) ratio and we converted the results to mean telomeric restriction fraction length (TRFL, kb) from a standard curve.This technique has been described by Cawthon, RM et al. [12].

Statistical analysis
Telomere length was compared with controls matched by age and sex using a Student's T-test.We used an F-test to compare factors associated with telomere length.The correlation coe cient between the telomere length and each parameter value was calculated using Pearson's correlation.The signi cance level was set at a P value < 0.05.

Compliance with ethical standards
Informed consents were completed by all patients before enrollment.This study was approved by the Human Research Ethics Committee of Faculty of Medicine, Chiang Mai University.(study code: MED-2559-03967)

TRFL in TDT patients compared to control group
The mean TRFL of the TDT group was 6.11 (± 0.6) kb.The mean TRFL of the control group was 6.79 (± 0.84) kb.TDT patients had a statistically signi cant shorter TRFL compared to the control group (p < 0.0001).(Figure 1)

Factor associated with telomere length in TDT patients
The correlation coe cients were calculated by Pearson's method.Moderate negative correlations were seen between the mean TRFL and age (r = -0.443,p = 0.0001), Figure 2, and there was a positive correlation with pre-transfusion Hb level (r = 0.251, p = 0.044), Figure 3.
The multivariate analysis factor showed that telomere length shortening was associated with age (p < 0.0001) and pre-transfusion Hb level (p = 0.02).
Sixty-three patients received iron chelation therapy whereas 2 patients did not receive iron chelation because of severe adverse effects from iron chelators.We found that the type of iron chelation therapy resulted in a difference in mean TRFL.Two patients who had no iron chelation had a shorter mean TRFL (5.1 and 5.74 kb) compared to patients who received iron chelation, and patients who received deferoxamine as an iron chelator had the longest mean TRFL compared to other iron chelators (6.61 kb) (p = 0.008).
There was no correlation between telomere length and other laboratory parameters including reticulocyte count, ferritin level, cardiac T2*, liver iron concentration, NTBI, or hemolysis parameters such as reticulocytes and liver enzymes.(Table 2) The multivariate analysis showed that only the age (p<0.0001) and pre-transfusion Hb level (p=0.027) were signi cantly associated with TRFL.
No difference was found between telomere length and gender (p = 0.068), type of thalassemia (p = 0.87), and splenectomy (p = 0.36).The majority of the population of patients had a high LIC [LIC 7-15 mg/g 25 (36.8%),>15 mg/g 28 (41.2%)] and there was a trend of a shorter telomere length in the high LIC group but this was not statistically signi cant (p=0.6).

Discussion
Telomere length is a well-known biomarker of cellular senescence [13][14][15].Our study demonstrated that patients with TDT had a shorter telomere length when compared with healthy controls.The telomere shortening was associated with the age and pre-transfusion Hb level.This nding suggested that patients with severe thalassemia exhibited accelerated telomere shortening.There was no statistically signi cant correlation between telomere length and other factors such as sex, reticulocyte count, ferritin, cardiac T2*, LIC, NTBI, and hemolysis parameters.
A recent study in patients with beta-thalassemia/HbE disease showed that accelerated telomere shortening was found only in patients who had severe clinical symptoms.Telomere length associated with aging was only found in patients with mild clinical symptoms but not in the groups with moderate and severe clinical symptoms.Reticulocyte count showed a correlation with telomere shortening.It was suggested that enhanced erythropoiesis in beta-thalassemia/HbE disease caused accelerated telomere shortening and demonstrated a disease-dependent trend that was not related to aging but is associated with reticulocytes [11].This nding was in agreement with our study in that the more clinical severe the symptoms, the shorter of telomere length.The difference in our study was the identi cation of pretransfusion Hb level as an indicator for disease severity, and not reticulocyte count.The explanation for this may be from the cross-sectional information as a spot reticulocyte count value may not represent the true reticulocyte count.In addition, some patients had other medical conditions that could in uence reticulocyte count such as endocrinopathy and heart disease [16,17].
Elevated body iron level leads to increased oxidative stress and long term exposure to oxidative stress leads to accelerated telomere shortening.Studies of hereditary hemochromatosis showed that elevated iron levels were associated with telomere shortening [18,19].However, we found no correlation between serum ferritin level and telomere length.This result was similar to the report in beta-thalassemia/Hb E disease [9], but differed from the ndings in hereditary hemochromatosis [18,19].This may be explained by the different population in the study and the method of body iron measurement.However, in our study, telomere length in patients who had LIC > 15 mg/g had a tendency to be shorter than those in the group of low LIC but the ndings were not statistically signi cant.Tissue iron may represent the body iron status of patients and is associated with accelerated shortening telomere length.
From our study, two patients who did not receive iron chelation due to severe adverse reactions to all chelation drugs had a shorter telomere length than other patients who received iron chelation.Patients who received deferoxamine had the longest telomere length.We hypothesized that the antioxidant effects of deferoxamine might be the protecting factor [20].Other previous studies showed that an antioxidant may help to protect against telomere length shortening [21,22].However, this nding about iron chelators needs to be interpreted carefully as patients who needed combination iron chelators may have had higher iron burden and experienced more chronic oxidative stress.Further studies should be conducted to evaluate the impact of iron chelation on telomere length.

Conclusions
Our study showed that patients with TDT had shorter telomeres.Telomere shortening in TDT thalassemia patients was an age-dependent process and was associated with greater clinical severity.Iron chelation may have a role to control telomere shortening.Further studies are needed for drugs targeting telomeres, for example testosterone that has been used for telomere elongation in patients with telomere diseases [23,24] and antioxidants that can protect telomeres from oxidative stress [21,22] for their role in improving the treatment outcomes in patients with TDT.

Tables
The telomere length of patients with TDT compared with controls.

Table 1
The clinical characteristics of enrolled patients

Table 2
Correlation coefficients and P values between mean telomere length and parameters in patients with *Denotes statistical significance at P < 0.05