Telomere Biology in Hematopoiesis and Stem Cell Transplantation

Abstract

Telomeres are long (TTAGGG)_n nucleotide repeats and an associated protein complex located at the end of the chromosomes. They shorten with every cell division and, thus are markers for cellular aging, senescence, and replicative capacity. Telomere dysfunction is linked to several bone marrow disorders, including dyskeratosis congenita, aplastic anemia, myelodysplastic syndrome, and hematopoietic malignancies. Hematopoietic stem cell transplantation (HSCT) provides an opportunity in which to study telomere dynamics in a high cell proliferative environment. Rapid telomere shortening of donor cells occurs in the recipient shortly after HSCT; the degree of telomere attrition does not appear to differ by graft source. As expected, telomeres are longer in recipients of grafts with longer telomeres (e.g., cord blood). Telomere attrition may play a role in, or be a marker of, long term outcome after HSCT, but these data are limited. In this review, we discuss telomere biology in normal and abnormal hematopoiesis, including HSCT.

Introduction

The replicative capacity of human cells is limited by the inability of DNA polymerase to fully replicate the linear ends of the DNA¹, a phenomenon called the “end-replication problem”. The physiologic solution to the end-replication problem was provided with the discovery of telomeres, the molecular structures at chromosome ends, and their maintenance enzyme, telomerase². Studies of telomere biology now range from a detailed understanding of telomeric structure, function, and regulatory mechanisms, to evaluating their role in aging, and human disease susceptibility, such as aplastic anemia, pulmonary fibrosis, heart disease, chronic inflammatory conditions, and cancer.

Telomeres shorten with successive cell division and, when a critical length is reached, cellular senescence is triggered. During the first weeks after hematopoietic stem cell transplantation (HSCT), the transplanted stem cells undergo extensive cell division to reconstitute the recipients’ new hematopoietic cells; this process has the potential to cause significant telomere attrition. In this review, we evaluate the 1) relationship between normal telomere biology and cell growth and maturation in the hematopoietic system, 2) bone marrow disorders associated with telomere biology abnormalities, and 3) telomere dynamics after HSCT, and their potential role in transplant outcomes. Finally, we discuss the clinical implications and directions for research on the role of telomere biology in HSCT.

Telomere Biology in Hematopoietic Cells

Human Telomeres:

Telomeres consist of long (TTAGGG)_n nucleotide repeats and an associated protein complex located at the end of the chromosomes. They are essential for maintaining chromosomal integrity by preventing chromosome end-to-end fusion³. Telomeres shorten with every cell division⁴, and therefore their length is a marker for cell aging, senescence and, replicative capacity. Telomere erosion in human cells represents a unique form of DNA damage, in which repair mechanisms are very limited⁵. When telomeres become critically short, cells become senescent and ultimately die. Alternatively, critically short telomeres may bypass senescence and continue to divide despite the presence of genetic instability. This is often associated with alterations in the p53 or Rb pathways, as observed in cancer cells⁶. Human studies show that telomeres shorten with age in almost all tissues, with the exception of the brain and heart (reviewed in⁷). Longitudinal studies suggest a positive relationship between the rate of telomere attrition and the individual’s baseline telomere length^8,9; individuals with the longest telomeres at baseline appear to have the highest attrition rate overtime. However, the mechanism of this differential telomere length regulation is unknown.

Telomeres are maintained mainly by the telomerase ribonucleoprotein complex and the shelterin protein complex. The catalytic unit of telomerase consists of an RNA template (TERC), a reverse transcriptase protein (TERT), and the dyskerin protein complex. Telomerase is expressed in embryonic and adult stem cells, and in highly-proliferative cells such as germline cells, skin, intestine, and bone marrow^10,11, but not in most other somatic cells¹². Approximately 80 to 90% of cancer cells maintain telomere length by over-expressing telomerase¹³. However, a subset of cancers elongate telomeres through an alternative lengthening mechanisms (reviewed in¹⁴). The shelterin complex consists of six protein subunits: telomeric repeat-binding factors 1, and 2 (gene names TERF1 and TERF2), TRF1-interacting nuclear factor 2 (TINF2), TERF2-interacting protein 1 (TERF2IP), TIN2-interacting protein 1 (ACD), and protection of telomeres (POT1)¹⁵. Shelterin is essential for normal telomere maintenance and function by inhibiting telomeric DNA damage response, regulating telomerase, and inhibiting telomeric homologous recombination¹⁶.

Human telomere length varies significantly between individuals. In human somatic cells, telomere length ranges between 2–15 kilobases (Kb)¹⁷, and is often longer in females than males¹⁸. In addition, intra-individual tissue differences in telomere length have been observed^19,20. However, within an individual, telomere length appears to be highly-correlated (i.e., individuals with longer telomeres in one tissue tend to have longer telomeres in other tissues)^21–25.

Several methods are available to measure telomere length (reviewed in^26,27). Terminal restriction fragment (TRF) measurement, based on Southern blots, is considered the gold standard. Other methods include: flow-FISH, which measures telomeres in leukocyte subsets; quantitative PCR (Q-PCR), which measures relative telomere length in extracted DNA as a ratio of telomere repeat copy number (T) to single gene copy number (S); and single telomere length analysis (STELA), a PCR method that measures chromosome-specific telomere length.

Telomere Dynamics in Hematopoietic Cells:

Proper telomere maintenance in hematopoietic stem cells (HSCs) is important because of the need for continuous replication of blood cells throughout life. Telomeres shorten rapidly after birth and during childhood. Early studies demonstrated the presence of longer telomeres in umbilical cord HSCs (CD34⁺CD38⁻) than in the same cell population in adult bone marrow²⁸ or in peripheral blood²⁹. Telomere shortening in peripheral blood mononuclear cells (PBMCs) during the first 3 years of life was found to be more than fourfold higher than that observed in adults during a period of 3 years³⁰. The profound rate of telomere attrition in the first years of life might reflect the expected rapid replication and expansion of hematopoietic stem cells early in life, followed by a decline in the rate of their replication thereafter³¹.

The presence of longer telomeres in umbilical cord blood (UCB) was not limited to stem cells, but was also observed in mature hematopoietic cells, including T-cells, B-cells, NK-cells, and granulocytes, compared with their adult counterparts²⁹. Additional studies suggest that lymphocytes have longer telomeres than granulocytes at birth, followed by a high attrition rate in the first year for both cell types. More prominent telomere loss was observed in lymphocytes after the first year of life, resulting in shorter lymphocyte telomeres (as measured by flow-FISH) in old age (>60 years) when compared with that of granulocytes³¹. Telomere length differences between lymphocyte subsets have also been reported: adult peripheral blood B-lymphocytes were shown to have longer telomeres than T-lymphocytes³². This was thought to be due to differences in the rate of cell division between B and T cells³³. In addition, CD8+ cells appear to have longer telomeres than CD4+ cells³⁴.

In support of the replicative senescence theory, longer telomeres were found in hematopoietic progenitor cells (CD34+) than in subsets of mature cells (naïve and memory CD4+ and CD8+, and granulocytes). Both CD4+ and CD8+ naïve T-cells have longer telomeres than corresponding memory cells³⁵, or effector cells³⁴. The longer telomeres in naïve cells might be the result of their higher replicative needs. However, one study found longer telomeres in memory B-cells than naïve cells³², a finding that still requires explanation. Overall, telomere length appears to be related to the replicative history of hematopoietic cells and their progenitors but potentially important deviations from this pattern have yet to be explained. Future studies geared toward understanding telomere biology in HSCs subsets will be important in understanding HSC development and their lifespan.

Telomere Biology and Bone Marrow Related Disorders

Dyskeratosis Congenita

The discovery that mutations in dyskerin (DKC1) caused X-linked recessive dyskeratosis congenita (DC)³⁶, and that these mutations resulted in abnormally short telomeres³⁷, were the springboard for studies of telomere biology in inherited disorders. DC is a complex disorder, characterized by the diagnostic triad of nail dystrophy, lacy reticular pigmentation of the neck and upper chest, and oral leukoplakia³⁸. Patients are at very high risk of bone marrow failure (BMF), myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). By age 30 years, 80–90% of DC patients develop severe aplastic anemia that does not respond to immunosuppressive medications³⁹; this is the main cause of death in those patients⁴⁰. Patients with DC are also at high risk of pulmonary fibrosis⁴¹, and cancer (11-fold higher than the general population)⁴².

DC is genetically heterogeneous, including X-linked recessive, autosomal dominant, and autosomal recessive forms. Approximately 25% of patients with DC have the X-linked form with mutations in DKC1. DC is also caused by inherited mutations in other critical components of the telomere maintenance pathway, including TERC (autosomal dominant), TERT (autosomal dominant and recessive), TINF2 (autosomal dominant), NOP10 and NHP2 (both autosomal recessive) and, the most recently discovered autosomal recessive gene, TCAB1⁴³. A mutation in one of these seven genes is identified in about 60% of patients with classic DC. Although all the DC genes have not yet been discovered, it is important to note that all patients with DC have very short telomeres for their age. Telomere length measurement in leukocyte subsets by flow cytometry with fluorescent in situ hybridization (flow-FISH) is sensitive and specific in distinguishing patients with DC from unaffected family members, and from patients with other inherited BMF syndromes⁴⁴. In another study using a different flow-FISH technique which determined relative telomere length in PBMC compared with a control cell line, all patients with DC had very short telomeres, but approximately 30% of patients with other BMF disorders also had telomere length ≤ 1st percentile⁴⁵. Clinical testing of telomere length by flow-FISH is now recommended for all patients with BMF, if testing for Fanconi anemia is negative^44,46.

Other Bone Marrow Failure Disorders:

Several studies have evaluated telomeres in patients with other inherited BMF syndromes e.g., Fanconi anemia, Diamond-Blackfan anemia, Shwachman-Diamond syndrome, and paroxysmal nocturnal hemoglobinuria⁴⁷. These studies suggest that some of these patients have shorter than average telomeres, but they are not as short as the very short telomeres which characterize DC. TL abnormalities in BMF syndromes other than DC have been observed primarily in granulocyte telomeres⁴⁴, an observation that might reflect the rapid turnover of hematopoietic cells in response to the bone marrow stress associated with marrow aplasia. Alternatively, the shorter telomeres in these disorders could also be due (in part) to their underlying genetic defect in DNA damage repair, particularly in Fanconi anemia and Shwachman-Diamond syndrome⁴⁸.

Similarly, patients with acquired BMF (i.e., aplastic anemia) also appear to have shorter telomeres than controls⁴⁹, but not as short as those observed in DC⁴⁵. Patients diagnosed with acquired BMF who have short telomeres are at high risk of relapse, clonal evolution, or mortality⁵⁰. The short telomeres in acquired BMF were believed to be secondary to the accelerated proliferation of the hematopoietic stem and progenitor cells in response to marrow hypoplasia⁴⁹. However, the presence of germline TERT or TERC mutations in approximately 10% of those patients⁴⁸ suggests the presence of occult DC cases among patients diagnosed with acquired BMF. Making this distinction is complicated by the observed phenotypic variability in DC, and made even more challenging because the causative gene is still unknown in around 40% of patients with DC. Incorporating telomere length measurement in the initial diagnostic studies for patients with acquired BMF is important, particularly because treatment options and outcomes are different for patients with inherited BMF (occult DC) than for those with truly acquired disease³⁹.

Myelodysplastic Syndromes

MDS are heterogeneous group of BMF disorders that are characterized by abnormal hematopoiesis, cellular dysplasia, and an increased risk of AML⁵¹. Patients with inherited or apparently acquired BMF are also at high risk of developing MDS⁴². MDS may be an intermediate stage between BMF and AML in these patients. In general, patients with MDS have shorter telomeres than healthy controls⁵². However, considerable variation in telomere length has been observed between patients with MDS⁵³. The presence of TERT or TERC mutations in some patients with familial MDS, in the absence of known clinical features of DC, further illustrates the possible presence of occult DC in these individuals⁵⁴. Germline mutations in TERT have been also reported in a subset of patients with AML⁵⁵ who do not meet clinical criteria for DC, again, suggestive of a clinical spectrum of diseases related to mutations in telomere biology genes.

Telomere dynamics in MDS may also be associated with disease characteristics and prognosis. Specifically, patients with isolated monosomy 7 were shown to have longer telomeres than patients with other cytogenetic subtypes, an observation that may be a consequence of telomerase over-expression in that patient subgroup⁵². Another study showed that MDS patients with the shortest telomeres have low hemoglobin concentrations, high percentages of marrow blasts, and a high incidence of cytogenetic abnormalities⁵⁶. Of note, telomerase activity did not correlate with clinical or prognostic parameters^53,57.

A link between short telomeres in MDS and leukemic transformation was suggested in two studies^56,58. However, in a study comparing telomere length and telomerase expression in MDS patients who did or did not progress to AML⁵³, short telomeres were observed in a similar proportion of patients in both groups. It seems that telomerase over-expression may be the driving force; this was mainly noted in patients with MDS who progressed to AML. MDS may be characterized by an early defect in telomere maintenance mechanisms, followed by a later telomerase over-expression which may be associated with malignant transformation⁵⁹.

Hematopoietic Malignancies

Telomere abnormalities have also been noted in patients with hematological malignancies^47,59–61. In most of the reviewed studies, telomere length and/or telomerase activity were measured in blood or bone marrow cells in patients’ samples and compared with healthy individuals; consequently, the patient cells analyzed were often a mix of malignant and benign cells. In general, most patients with acute or chronic leukemia have short telomeres and high telomerase expression, with some variation by disease subtype. In a study comparing telomere length and telomerase activity in hematopoietic progenitor cells (CD34+) collected from normal individuals and from patients with different types of acute and chronic leukemia, short telomeres were observed in AML and chronic myeloid leukemia (CML) specimens, but not in chronic lymphocytic leukemia (CLL). On the other hand, telomerase activity in AML specimens was significantly higher than that of the controls, a moderate elevation was observed in CML, and no change in CLL⁶²; this may be a consequence of telomere length variation in these diseases. In a recent study that directly compared telomere regulation in various acute leukemias, telomeres were the shortest in B-ALL, followed by T-ALL and then AML; short telomeres were also associated with higher telomerase expression in these patients⁶³. In CML, a few studies compared malignant hematopoietic stem cells (BCR-ABL CD34+) with benign stem cells^64,65 and confirmed the previously-observed telomeric dysregulation in the form of short telomeres and elevated telomerase activity.

Telomere biology in hematopoietic malignancies has been also associated with clinical characteristics and outcome. Higher telomerase levels have been associated with relapse, disease progression, and worse survival in adults with acute or chronic leukemia^66–68, but not in children with ALL⁶⁹. In CML, telomere shortening has been observed in serial peripheral blood samples collected from patients who progressed from chronic to blast phase⁷⁰, and it was associated with faster time to progression in another study⁷¹. In addition, patients with CML in blast phase were reported to have significantly higher telomerase activity compared with those in the chronic phase⁷². High levels of telomerase in CML blast phase was also associated with additional cytogenetic changes and microsatellite abnormalities, indicating the development of genomic instability⁷³. It is important to note that cytotoxic chemotherapy can shorten telomeres, mostly because of the hematopoietic cell regeneration required for recovery from drug-induced marrow suppression, while the use of the tyrosine kinase inhibitor, imatinib, in patients with CML was found to elongate previously shortened peripheral blood telomeres. This phenomenon was suggested to be due to the shift from Philadelphia chromosome positive (Ph+) to Ph- cells in patients’ blood in response to treatment⁷⁴. A subsequent study of the effect of imatinib on telomere length found that it significantly down-regulatedD telomerase activity in primary leukemic cells through a direct action on TERT transcription⁷⁵.

Role of Telomeres in HSCT

HSCT has become the treatment of choice for many hematologic malignancies, as well as other inherited and acquired bone marrow and immune system disorders. Annually, more than 30,000 autologous, and 15,000 allogeneic transplantation are performed worldwide⁷⁶. Details about HSCT history, methods, early and late complications, as well as outcomes in selected diseases were reviewed in⁷⁶. HSCT recipients receive a preparative regimen of chemotherapy with or without irradiation to eradicate malignant clone and/or induce immunosuppression prior to HSCT which permits engraftment. HSC preparations are then infused into the transplant recipient. The main sources of stem cells in HSCT are bone marrow, peripheral blood, and UCB. The infused stem cells undergo acute and rapid replication and proliferation (at much greater than normal rates) to reconstitute the recipient’s hematopoietic system. In most cases, peripheral blood cell counts and bone marrow cellularity normalize within 2–6 months after HSCT⁷⁷. However, donor hematopoietic progenitor cell function remains deficient in the recipient for several years after transplantation^78,79. Specifically, bone marrow CD34 cells from transplant recipients (at a median of 3 years post-transplant) have lower colony scores and cytokine concentrations in erythroid burst-forming units (BFU-Es), erythroid and myeloid colony forming units (CFU-Es), and colony forming units-granulocyte macrophage (CFU-GM)⁷⁸.

Proper telomere function is important in HSCT because telomeres shorten with each cell division; the donor HSCs undergo massive proliferation after infusion in order to achieve engraftment (Figure 1). The interplay of cell division and telomere shortening thus has potential to affect clinical outcome. A limited number of studies have evaluated the short and long-term telomere length dynamics after HSCT by comparing the donors’ blood or marrow telomeres to those within the cells transplanted into the recipients. In addition, some studies have evaluated the effect of telomere attrition on hematopoietic reconstitution, or on factors contributing to post-transplant telomere attrition.

Figure 1:

A model for telomere dynamics and outcomes after hematopoietic stem cell transplantation (HSCT). Relative telomere length is shown on the Y-axis and relative time is on the X-axis. After infusion of donor hematopoietic stem cells (HSC), rapid telomere length attrition occurs then levels off to relatively normal, age-related rates. Critically short telomeres may occur earlier in HSCT recipients.

Telomere Dynamics After HSCT

After HSCT, telomeres in the recipient are typically shorter than in their matched donors⁸⁰. This pattern has been reported in both autologous and allogeneic HSCT, suggesting that this effect is due to rapid HSC division on telomere length. In autologous transplantation, telomere shortening was most prominent in elderly recipients who had undergone several courses of pre-HSCT chemotherapy. This suggests that pre-transplant therapy which affects the bone marrow stroma and its micro-environment in the host may play a role in post-transplant telomere length regulation. In recipients of allogeneic transplantation, telomeres were shorter in those who received HSC from elderly donors than in those recipients who had younger donors⁸¹. This primarily reflects the fact that older donors have shorter telomeres because of the normal age-related telomere attrition in healthy individuals. Of note, one study suggested that the use of nonmyeloablative conditioning regimens did not prevent post-transplant telomere shortening in the recipient⁸².

Data from longitudinal studies with repeated telomere length measurements of the same individual^83,84 showed rapid telomere length attrition in the first 12 months after HSCT in recipients of bone marrow grafts. This is followed by telomere shortening rates that were similar to the expected age-related decline seen in controls. The reported annual rates of telomere shortening vary based on the study and method of telomere length measurement. It has been estimated that post-HSCT telomere attrition rate in the first year is 537 ± 155 base pairs (bp), compared with an annual attrition rate of 221 ± 81 bp thereafter⁸⁴. Looking more closely, Thornley and colleagues⁷⁷ reported significantly shorter telomeres in the recipients at engraftment, 6, and 12 months post-transplant compared with donor telomere length before bone marrow harvest or after mobilized peripheral blood collection. Evaluation of the changes in recipients’ telomere length over the same time period did not find significant telomere attrition. This suggests that the maximum telomere attrition occurs within weeks after HSC infusion. These studies illustrate the utility of telomere length as a biomarker of the excessive cellular replication which occurs in order to achieve engraftment.

A similar, but less profound, pattern of post-HSCT telomere shortening has been observed in recipients of peripheral blood stem cell transplants (PBSCT). Compared with donors, the average neutrophil telomere shortening for bone marrow transplant (BMT) recipients was 321 ± 90 base pairs (bp) vs. 296 ± 127 bp for PBSCT recipients. A similar pattern was observed for telomere shortening in T cells (289 ± 133 bp for BMT recipients’ vs. 113 ± 379 bp for PBSCT recipients)⁸⁴. At one year after transplantation, and in contrast to bone marrow stem cell transplantation, telomere length in PBSCT recipients was comparable to that of the donors in one study⁸⁵. This observation needs to be confirmed. If true, this could be due to the presence of a higher number of HSC in PBSCs, with fewer cell divisions required to achieve engraftment^86,87. However, the correlation between the number of infused HSCs and recipients’ telomere length was inconsistent between several studies^81,84,88.

UCB is an important source of stem cells and their use in HSCT is growing. Telomeres in UCB cells are longer than those in other HSCT sources because they are acquired at birth, when telomeres are the longest and normal age-associated telomere attrition has not yet occurred. One study showed that the degree of post-transplant telomere attrition in UCB and PBSCT recipients at one-year was comparable (difference between donor and recipients TL ranged between 0.3–0.6 Kb in UCB, and 0.4–0.7 Kb in PBSC). As expected, the absolute TL was longer for UCB HSCT recipients^89–91. In support of this finding, Ruella and colleagues reported that in both autologous and allogeneic transplants, post-transplant (median= 14 months) bone marrow telomere length reflected that of the grafted cells⁹².

It is conceivable that activation of telomerase may help in controlling post-transplant telomere attrition. For example, antigen stimulation of T cells derived from serially transplanted HSCs resulted in telomere elongation in wild type but not in telomerase deficient mice, suggesting that telomere elongation after antigen stimulation is telomerase-dependent⁹³. A subsequent study showed that⁹⁴ serially transplanted HSCs from telomerase deficient mice have lower replicative capacity (rounds to exhaustion=2 vs. 4 or more), and higher rate of telomere attrition (2-fold increase), when compared with that transplanted from wild type.

Studies of telomere length changes in long-term HSCT survivors are important in understanding the replicative capacity of the transplanted HSCs. This will become more important as transplant recipients survive for longer because there could be late graft failures if the donor cells have reduced replicative capacity due to shorter telomeres. Baerlocher and colleagues⁹⁵ compared post-transplant peripheral blood telomere length in leukocyte subsets of long term HSCT survivors and their donors at the same time intervals and found that the recipients’ telomeres were shorter in all leukocyte subsets studied (granulocytes, naïve and memory T cells, B cells and natural killer cells). In the same study, shorter telomeres were associated with the presence of chronic graft versus host disease (GvHD) and receiving graft from a female donor. On the other hand, a small, but not statistically significant, telomere length difference between long term transplant survivors (median follow up was 18 years) and their donors was observed in another study⁹⁶. The inconsistent results between the two studies might be related to the differences in recipient and donor factors, such as the donors’ gender and the presence of GvHD in the recipients.

While the literature on telomere dynamics after HSCT is growing, it is important to note that most of the available studies were cross-sectional in design, and included a very small number of participants. Survival bias may be of concern in studies that evaluated telomere length in long-term survivors. It is possible that transplant recipients with the shortest telomeres died early after transplantation and therefore were not represented in most of the available studies. A more comprehensive investigation is needed to answer some of the unsolved questions.

Factors Influencing Telomere Dynamics After HSCT

Telomere Dysfunction in the Recipient:

TERC knockout mice (TERC^−/−) provide an experimental model in which to study the involvement of telomere biology in aging, cancer development, cardiovascular diseases and others. In general, TERC^−/− mice lack detectable telomerase activity because they lack the telomerase RNA template, TERC. However, variability across generations has been noted. Telomere length in these mice shortens in subsequent generations by approximately 5 Kb. Phenotypic abnormalities do not appear until the 3rd generation⁹⁷, most likely because mice telomeres are very long compared with humans (40–80 Kb vs. 10–15 Kb, respectively)⁹⁸, and knocking out telomerase appears to cause gradual telomere shortening, at least in mice. Aging TERC^−/− mice showed impaired bone marrow stromal cell function, with altered cytokine expression, that contribute to impaired engraftment of transplanted wild-type HSCs⁹⁹. Additionally, HSCT did not rescue thymic function in these mice¹⁰⁰. Taken together, these observations suggest a role of telomere biology on HSCT outcomes. Studies involving patients with DC and related telomere biology disorders provide a very important opportunity to better understand these interactions.

BMF is very common in patients with DC, and HSCT is the only option for cure. Large, prospective studies of HSCT outcomes in patients with DC have not yet been conducted, but compilations of case reports have yielded important insights. Collectively, severe organ toxicity, particularly in patients with DC who had received a myeloablative regimen, was the most common cause of death⁴⁰. This observation highlights the importance of telomere maintenance on tissue repair, particularly after the exposure to DNA damaging agents in preparative regimens. In agreement with published reports, and in a case series of 11 patients with DC who were transplanted in different centers across the U.S., only 4 survived⁴². Of note, one of the survivors developed severe pulmonary fibrosis 8 years post-transplant. Late post-transplant graft failure has been reported in several recipients who had very short blood telomere lengths after engraftment¹⁰¹.

A recent report of a DC-specific non-myeloablative regimen, which included cyclophosphamide (single dose), fludarabine, alemtuzumab, and total body irradiation (200cGy) reported a favorable short term outcome in a small number of patients (67% survived with a median follow-up of 26.5 months), but long-term outcomes are unknown¹⁰². In DC, it is especially important to exclude or reduce the dose of agents with known pulmonary and liver toxicity, in order to minimize the risk of organ fibrosis. Identifying patients with DC prior to HSCT, through telomere length testing, is important not only for selecting the least toxic/most effective transplant regimen, but also for appropriate donor selection decision-making. Clinically normal family members of DC patients are at some risk of being silent carriers of the same genetic disorder, which would make them ineligible as donors. For example, an adult patient who received HSCT for aplastic anemia from his apparently-healthy matched sibling had a complicated post-transplant course, in the form of delayed engraftment with long-lasting neutropenia and died of sepsis approximately 6 month after transplantation¹⁰³. The sibling donor – who had no clinical evidence to suggest DC when marrow stem cells were collected - was subsequently found to have the same TERC mutation as the recipient. Using telomere length as a diagnostic tool for DC in this setting is particularly important for patients who do not have a mutation in one of the known genes, and as a DC screening test for patients who do not present with the typical clinical triad.

Donor age:

Increasing donor age is associated with reduced overall and disease-free survival after HSCT. The presence of shorter telomeres in older donors suggests that their HSC may not have the same replicative capacity as HSC from younger donors. In a large follow-up study of 6,978 recipients of unrelated donor bone marrow HSCT¹⁰⁴, statistically significant trends, based on increasing donor age 18 to 30 years, 31 to 45 years, and more than 45 years), were noted in five-year overall survival (33%, 29%, and 25%, respectively), cumulative incidence of severe acute (30%, 34%, and 34%), and chronic (44%, 48%, and 49%) GvHD. In addition, transplant recipients who received HSC from young donors were at lower risk of developing obstructive lung diseases¹⁰⁵, including bronchiolitis obliterans¹⁰⁶, secondary graft failure after initial engraftment¹⁰⁷, and B-cell lymphoproliferative disorder¹⁰⁸. The relatively limited cellular replicative capacity of older donor cells has also been reflected in solid organ transplantation. Donor age was found to be the strongest predictor for long term outcomes in renal transplantation, and was significantly associated with markers of cellular senescence in transplanted kidney^109,110. Since telomere length shortens with age, it provides a biological explanation for the observed relationship between donor age and transplant outcome.

Inflammation and infections:

During the immediate course following HSCT, and even years after, transplant recipients are at risk of numerous complications, most commonly infections and GvHD. Long-term HSCT survivors who had chronic GvHD were more likely to have short telomeres in white blood cell subsets than those who did not⁹⁵. This association could be explained by the fact that telomeres are sensitive to oxidative stress and chronic inflammation. Short telomeres have been reported in many chronic inflammatory conditions such as rheumatoid arthritis¹¹¹, systemic lupus erythematosus¹¹², ulcerative colitis¹¹³, atherosclerosis, and diabetes¹¹⁴. In addition, telomere shortening is associated with genetic variation in oxidative stress genes¹¹⁵, and genes regulating mitochondrial reactive oxygen species production¹¹⁶.

However, the observed association between short telomeres and chronic GvHD could also be explained by the use of immunosuppressive medication. In some instances, this effect seems to be agent-specific. Exposure of renal cells in vitro to cyclosporine A induced cellular senescence and shorter telomeres¹¹⁷. Exposure of human T-lymphocytes to cortisol was associated with a significant reduction in telomerase activity¹¹⁸. In a yeast study, TOR (Target of Rapamycin) inhibition suppressed cell death caused by inactivation of yeast telomere binding protein Cdc13p, but did not change telomere length¹¹⁹. It has been suggested that short telomeres in patients with GvHD correlates with karyotypic abnormalities and malignant transformation¹²⁰.

Infections are common in HSCT and comprise one of the main causes of death in transplant recipients¹²¹. Chronic infections outside of the transplant setting, such as cytomegalovirus (CMV) and hepatitis C (HCV), have been associated with short lymphocyte telomeres^122,123. Chronic GvHD is an important risk factor for infectious events in HSCT recipients¹²⁴, an association that is thought to be related to the immunosuppressive state that results from GvHD. HSCT recipients who have GvHD, or were exposed to repetitive infections, had short telomeres in isolated effective memory cells (CD27⁻ CD4⁺), compared with healthy controls¹²⁵. Further evidence from mouse models highlighted the interaction between telomere function and immune responses. In late generation TERC^−/− mice, telomerase deficiency was associated with impaired antibody-mediated immune responses¹²⁶. In wild-derived house mice accelerated telomere shortening was noted in response to antigenic stimulation, and repeated infection¹²⁷. Taken together, these data suggest that recurrent infections shorten telomeres, which subsequently limit T cell proliferative capacity, and can impair the response to infection. Both chronic GvHD and inflammation may be important components of this association.

Lifestyle and behavioral factors:

Epidemiologic studies in non-HSCT populations suggest that various lifestyle factors, such as smoking, diet, and exercise, are associated with telomere length^128–130. In addition to the medical complications related to HSCT, the same factors that affect telomere biology in the general population may also be important in HSCT recipients. For example, at least two studies suggest that a healthier lifestyle, defined by the presence of a lower BMI, more exercise, no smoking, and higher intake of fruits and vegetables was associated with longer telomeres¹³¹. Higher intake of dietary fiber and lower polyunsaturated fatty acid were associated with longer telomeres in the Nurses’ Health study¹²⁸. Several studies have reported an association between smoking^130,132,133 and short telomeres. Longer telomeres are associated with increased physical activity¹²⁹, long-term use of hormonal replacement therapy¹³⁴, less stressful lifestyle¹³⁵ and healthier diets¹²⁸.

There is accumulating evidence of a negative relationship between telomere length and psychosocial stress, a common observation in transplant recipients and their family members particularly in the peri-transplant period¹³⁶. Shorter telomeres have been associated with mood disorders¹³⁷, self reported stress¹³⁵, and early social deprivation¹³⁸. Additionally, levels of perceived stress and its duration were associated with lower telomerase activity, and higher oxidative stress index¹³⁵. In a randomized controlled trial, a 3-month mediation program was associated with higher telomerase activity, believed to be mediated by decreased level of stress in the experimental group compared with control¹³⁹. Additional studies have suggested that lifestyle modified the relationship between telomere length and atherosclerosis: healthier lifestyles in the form of more social support, lower meat consumption, as well as high fruit and vegetable consumption attenuated the risk of atherosclerosis in individuals with short telomeres¹⁴⁰. Figure 2 provides examples of these environmental and lifestyle factors that might also affect the telomere biology of the donor cells in a HSCT recipient.

Figure 2:

Factors that may affect telomere length in HSCT recipients. Telomeres are sensitive to oxidative stress and DNA damage. Telomere length is associated with many factors, including age, genetics, and environmental exposures.

Abbreviations: HSCT, hematopoietic stem cell transplantation; GvHD, graft versus host disease

Telomere Biology in Induced Pluripotent Stem cells (iPSCs)

The creation of induced pluripotent stem cells (iPSCs) has the potential to create a source of functioning differentiated cells, including hematopoietic cells. Similar to embryonic stem cells, iPSCs have self renewal, expansion, and differentiation capabilities^141–143. This leads to the possibility of reprogramming patient-specific cells for use in HSCT. The current advances of human iPSCs in blood engineering were recently reviewed in¹⁴⁴. Telomere dynamics in iPSCs and during stem cell differentiation are an important measure of cellular replicative capacity. Using iPSCs generated from mouse embryonic fibroblasts, Marion and colleagues¹⁴⁵ showed that generated cells’ telomere length and telomere heterochromatin markers were similar to embryonic stem cells. Of interest, the generation of iPSCs from telomerase deficient mice was severely impaired; an observation that was recently confirmed in patients with DC phenotype caused by defects in telomerase complex genes (TERT, DKC1, or TCAB1)¹⁴⁶. Telomeres were found to be significantly longer in iPSCs derived from human neonatal fibrblasts compared with the original cells¹⁴⁷. They also progressively shortened after differentiation and reached a length similar to that of the differentiated fibroblasts. Telomere elongation in human iPSCs was associated with telomerase activation followed by telomerase down-regulation after differentiation. Of importance, the observed telomere elongation in iPSCs was independent of original cell age, where iPSCs derived from either old or young donors restored telomere length to that similar to embryonic stem cells. This observation was consistent in mice¹⁴⁵ and human experiments¹⁴⁸. Details about telomere restoration and dynamics during nuclear reprogramming have been recently reviewed in^149,150.

Does Telomere Biology Have A Role In HSCT Outcome?

Survival after HSCT has shown significant improvement during recent decades, primarily due to advances in treatment modalities and supportive care. However, HSCT survivors are still at high risk of developing serious complications, including GvHD, serious infections, pulmonary complications, organ failure, and cancer¹⁵¹. The similarities between some of the post-HSCT complications and those associated with short telomeres in the general population suggest a role for telomere maintenance in short and long-term outcomes after HSCT.

Studies of the relationship between telomere defects, including short telomeres, and human diseases demonstrated an association with BMF, pulmonary and liver fibrosis, cancer, cardiovascular diseases, chronic inflammation, and premature aging⁴⁷. In addition, epidemiological studies in the general population, show associations between short telomeres and excess risk of cancer, overall¹⁵², and in certain anatomic sites such as bladder, kidney, lung, head and neck¹⁵³, and breast¹⁵⁴. A recent meta-analysis¹⁵⁵provided evidence of a modest association between short telomeres, comparing the shortest with the longest quartile, and cancer risk (OR=1.96, 95% CI=1.37–2.81). Data on the effect of post-HSCT telomere dynamics on long-term transplant outcomes are very limited. Late graft failure after primary engraftment has been suggested to be associated with short telomeres in 2 case reports¹⁰¹. A relationship between short telomeres and chronic GvHD has also been reported⁹⁵, but it is not clear whether this relationship is specific to certain types of GvHD, such as fibrotic, or hematologic. Chronic GvHD in HSCT recipients is associated with elevated risk of second cancers¹⁵⁶, an association that has been explained, in part, by the long term use of immunosuppressive medication in those patients¹⁵⁷. The relationship between GvHD in HSCT recipients and short telomeres⁹⁵, might explain, at least in part, cancer development in those patients; a hypothesis that needs to be tested.

A comprehensive assessment of the role of telomere length and its relationship to HSCT transplantation outcomes is needed. A longitudinal study with a large sample size would be essential to assessing changes of telomere length over time, starting with telomere length in infused cells, and to evaluating the effect of telomere dynamics on HSCT outcomes. If a relationship exists, it will be important identify factors affecting this dynamics, particularly those that can be modified, such as donor selection, type of preparative regimen, type of immunosuppressive medication, and/or recipient lifestyle.

Conclusions:

In healthy individuals, the gradual telomere loss in hematopoietic cells with age has been suggested to contribute to declines in hematopoiesis, immunosenescence, and malignant transformation. In HSCT, the telomeres of the donor, experience significant attrition once infused into the recipient, particularly during engraftment and in the first post-transplant year. The rate of telomere attrition does not seem to differ significantly by stem cell source. However, UCB transplants were associated with longer telomeres than other sources, mostly because UCB have longer telomeres at the beginning. Telomere length in the blood cells of HSCT donors is a biomarker of both their biological and physiological age, and may be correlated with long-term outcomes after HSCT. We hypothesize that transplant recipients who receive HSCT from a donor with short telomeres, or who experience severe telomere attrition as a result of transplant-related complications, may be at high risk of graft failure, second malignancies, and other complications. Large, prospective studies are needed to test this hypothesis.

Practice Points:

• Incorporating telomere length measurement into the initial evaluation of patients with aplastic anemia (BMF) is currently the most cost-effective way to identify patients with BMF patients who have a telomere biology disorder.

• Identifying patients with classic or occult dyskeratosis congenita (i.e., a telomere biology disorder) prior to HSCT is important transplant regimen decision-making (with consideration of non-myeloablative regimens).

• Telomere length should be measured in potential HSCT donors who are related to a recipient with short telomeres because they may be clinically silent but carry the same genetic abnormality. This may result in graft failure.

• In HSCT for DC (classic and occult) and related telomere biology disorders, it is important to exclude or reduce the dose of agents with known pulmonary and liver toxicity, to avoid potentially life-threatening end-organ fibrosis.

Research Agenda:

• Identify the long term outcome of HSCT for patients with telomere biology disorders

• Determine the best conditioning regimen for patients with telomere biology disorders

• Determine the role of donor telomere length in recipients’ outcomes after HSCT