Telomere Length

Shortening of telomeres in recipients of both autologous and allogeneic hematopoietic stem cell transplantation

Article metrics


Telomere length of peripheral blood mononuclear cells (PBMCs) from 23 autologous HSCT patients ranging from 4 to 61 years old, and 46 allogeneic HSCT recipients from 6 to 52 years old were studied to confirm whether excessive shortening of telomeres is associated with HSCT. After autologous HSCT, telomere length of PBMCs ranged from 6.8 to 12.0 kb. The comparison between transplanted PBMCs and PBMCs after autologous HSCT showed shortening by up to 1.9 kb (mean ± s.d.: 0.64 ± 0.50 kb). There was a difference between autologous HSCT patients and normal volunteers in the slopes of regression lines. After allogeneic HSCT, telomere length of PBMCs ranged from 6.8 to 12.0 kb. Telomeres of recipients were up to 2.1 kb (0.60 ± 0.468 kb) shorter than those of donors. The slope of regression lines for allogeneic HSCT patients and normal volunteers were parallel. Although all patients were transplanted with more than 2.0 × 108 cells/kg, telomere length did not correlate with the number of transplanted cells. There was no significant correlation between telomere length and recovery of hematological parameters. However, three patients with an average telomere length of 6.8 kb after HSCT took a longer period to reach the normal hematological state. Taken together, these data suggest that most HSCTs are performed within the biological safety range of telomeres, while the patients who have telomeres shorter than 7.0 kb after HSCT should be observed carefully for long-term hematopoiesis and the occurrence of hematopoietic disorders. Bone Marrow Transplantation (2000) 25, 441–447.


Telomeres, composed of protein and 2–15 kb long tandem repeat sequences (5′-TTAGGG-3′)n at the end of chromosomes, protect from end-to-end fusion and exonucleotic degradation.12 Telomeric DNA is shortened by 50–100 bp with each somatic cell division, because of incomplete end-replication by DNA polymerase and poor telomerase activity, which adds telomere sequences to the end of chromosomes.34 The telomere hypothesis proposes that a critical shortening of telomeres leads to cessation of irreversible cell division. Indeed, telomeres play the role of a mitotic clock, a good indicator of replication history and of the residual replication potency of cells.5678

Hematopoietic stem cell transplantation (HSCT) is applied for treatment of various malignant and non-malignant diseases. Recovery of the hematopoietic system after HSCT imposes numerous replication cycles on hematopoietic stem cells (HSCs). Such enforced excessive divisions of HSCs may cause excessive shortening of telomeres in their descendant cells, although HSCs possess telomerase activity. Indeed, recent studies including ours have demonstrated that HSCT induced less than 1.5 kb loss of telomeres.91011 The telomeric shortening that occurs during transplantation is the same as the amount that would be seen in 15–20 years of steady-state hematopoiesis in normal individuals.12 This may be due to telomerase activity in HSCs, which is not sufficient to compensate completely for the reduction in telomeres along with enforced hematopoietic recovery. These studies have been mainly done in younger patients who underwent HSCT. No data is available on the effects of HSCT for elderly adults.

The present study addressed telomere length of peripheral blood mononuclear cells (PBMCs) in patients who underwent HSCT, including elderly patients and elderly donors. We found that the shortening of telomeres following HSCT was up to 2.1 kb. Telomere length was inversely correlated to age. The effect of HSCT on telomeres was much more prominent in elderly patients who underwent several courses of cytoreductive therapy in autologous HSCT and in patients who received transplants from elderly donors in allogeneic HSCT.

Patients and methods


Sixty-nine patients who underwent HSCT from 1983 to 1998 at Jikei University Hospital, Tokyo Metropolitan Komagome Hospital, Tokai University Hospital, and Niigata University Hospital were studied. Clinical characteristics of the patients are listed in Tables 1 and 2. Conversion to donor blood type in allogeneic HSCT was confirmed by studying one or two of the following: sex chromosomes, types of red blood cells and types of HLA. The mean follow-up period after HSCT was 3.2 years in the autologous HSCT group and 2.6 years in the allogeneic HSCT group. Donor age ranged from 7 to 70 years old (median 28 years). All patients received cytoreductive therapy followed by HSCT. Hematological parameters of recipients were studied after 1 year from HSCT. Some cases were previously presented as indicated in Tables 1 and 2.11

Table 1  Characteristics of autologous HSCT patients
Table 2  Characteristics of allogeneic HSCT patients

Analysis of telomere length

All samples were obtained from peripheral blood or bone marrow in accordance with the ethical guidelines of each hospital under informed consent from patients, donors or parents. Mononuclear cells were isolated by the gravity centrifugation method, and high molecular weight DNA was obtained by standard phenol/chloroform extraction. Then, 2 μg of HinfI-digested DNA was size-fractionated by 0.7% agarose gel electrophoresis. After electrophoresis, the size-separated DNA was transferred to a Hybond N filter (Amersham, Amersham, UK) by the vacuum transfer method (BC-650 Biocraft, Tokyo, Japan), and cross-linked with the filter by exposure to ultraviolet light. Filters were hybridized with telomeric probe (TTAGGG)4, labelled with digoxigenin according to the manufacturer's instructions (Boehringer, Mannheim, Germany). Hybridization was carried out at 68°C for 12 h in the hybridization solution (5 × SSC, 0.02% SDS, 0.1% sodium-lauroylsarcosine, and 1% blocking reagent (Boehringer Mannheim)). The filter was washed twice at 50°C in the washing solution (0.1 × SSC, 0.1% SDS) and rinsed in the recommended blocking reagent. The hybridized probe was detected by the chemiluminescence method according to the manufacturer's recommendations (Tropix, MA, USA). Filters were exposed with Fuji New RX film (Fuji Film, Tokyo, Japan). The films were then scanned with a Canoscan 600 (Canon, Tokyo, Japan) and the average telomere length was analyzed using NIH image software. Briefly, the strongest density peak was tentatively taken to be the mean telomere length.91113

Statistical analysis

Linear regression analysis, paired t-test and Mann–Whitney U test were performed by using Stat View4.5 software (Abacus Concept, CA, USA).


Telomere length in normal volunteers

Previous data demonstrated that the reduction of telomeres is biphasic.14 After 4 years of age, the rate of reduction becomes slower. PBMCs from 39 normal volunteers from 4 to 70 years old were studied. As shown in Figure 1, the shortening of telomere length was proportional to aging at 34 bp per year (r = −0.691, P < 0.0001). This was similar to previous studies,81015 but differed slightly from our previous data,11 owing to different numbers and ages of volunteers.

Figure 1

Telomere length of PBMCs in normal volunteers and HSCT patients. In a and b the broken line denotes the regression line of normal volunteers. The equation is T = 10.778 − 0.034 × A, where T = telomere length (kb), and A = age (years) (r = −0.691, P < 0.0001). The scattergram of normal volunteers is not shown. (a) Autologous HSCT patients. The scattergram of telomere length vs age was demonstrated. The bold line demonstrates the regression of patients. The equation is T = 11.118 − 0.055 × A (r = −0.713, P < 0.0001). (b) Allogeneic HSCT patients. The scattergram of telomere length vs age is demonstrated. The bold line demonstrates the regression of patients. The equation is T = 10.122 − 0.034 × A (r = −0.387, P < 0.0001).

Telomere length in PBMCs of the patients

The mean telomere length of PBMCs from patients who underwent autologous HSCT was 9.70 kb (range, 6.8–12.0 kb) (Table 1). Among autologous HSCT recipients, 11 paired samples of transplanted PBMCs (case 15: bone marrow cells) and PBMCs after hematopoietic reconstitution (ΔTelauto) could be compared. The ΔTelauto was 0.60 kb shortening on average during an observation period of 0.3 to 5.3 years. These telomere losses are by the same amount as expected in 15–20 years of steady-state hematopoiesis in normal individuals. Next, autologous HSCT patients were compared with controls deduced from Figure 1. No significant differences in telomere length were found, when all generations were studied, but a significant difference was observed in a group of older than 40 years (Mann–Whitney U test; P = 0.01). Therefore, the steep decline of regression line of autologous HSCT patients was biased by the over 40 year group (Figure 1a). Case 16 was a non-Hodgkin's lymphoma patient who underwent autologous BMT with PBSC after sequential chemotherapies. The telomere length in this patient was 10.6 kb, and ΔTelauto showed the greatest reduction (1.9 kb) in the autologous HSCT group (Table 1). Cases 19 and 20 were elderly non-Hodgkin's lymphoma patients who received BMT and BMT with PBSC after cytoreductive chemotherapies, respectively (Table 1). They had the shortest telomere length (6.8 kb) in the autologous HSCT group which would correspond to an age of 100 years according to the equation for normal controls (Figure 1).

The mean telomere length of allogeneic HSCT recipients was 9.30 kb (range, 6.8–12.0 kb) (Table 2). The telomeres of recipients after hematopoietic recovery were shorter than those of donors by 0.60 kb on the average (ΔTelallo) (n = 15) (Table 2). This difference between donors and recipients was statistically significant (paired t-test; P = 0.0002). The regression line for allogeneic HSCT recipients was virtually parallel to that of controls (Figure 1b). We next compared telomere length between patients and normal controls. The telomeres of allogeneic HSCT recipients were significantly shorter than those of putative donor controls for all ages (Mann–Whitney U test; P = 0.04) but not those of putative recipient controls (P = 0.09). Case 74 was a 50-year-old chronic myelogenous leukemia patient who underwent PBSCT from his 70-year-old mother, and his mean telomere length was 6.8 kb. This case had the shortest telomere length in the allogeneic HSCT group (Figure 2b). Also, comparison of telomeres between donor and recipient revealed 2.1 kb shortening in the recipient, which would correspond to 20 cell replications.

Figure 2

The dynamic changes of telomere length with HSCT. Southern blot analysis of telomere length was performed as mentioned in the Materials and methods. (a) Case 19 who underwent autologous BMT plus PBSC. 1: Transplanted BM cells (8.7 kb); 2: transplanted PBMCs (8.0 kb); 3: PBMCs after recovery of hematopoiesis (6.8 kb). (b) Case 74 who underwent allogeneic PBSCT. 4: transplanted BM cells (8.9 kb); 5: donor PBMCs (8.8 kb); 6: recipient PBMCs after recovery of hematopoiesis (6.8 kb).

Relationship between telomere length, age, and number of infused cells

As demonstrated in Figure 1a, telomeres become shorter with aging. Indeed, telomere length after autologous HSCT correlated with age (r = −0.713, P < 0.0001) (Figure 1a). However, ΔTelauto did not correlate with patient age (r = 0.165, P = 0.608).

Allogeneic HSCT was much more under the influence of ageing than autologous HSCT. We found a reverse correlation between donor age and ΔTelallo (r = 0.811, P = 0.0002) (Figure 3a). A previous study demonstrated that the number of transplanted cells affects the recipient's telomere length.9 Our results were also analyzed for this. Unlike previous data, the number of infused cells in allogeneic HSCT was independent of telomere length and ΔTelallo (r = 0.181, P = 0.5742), although all patients in our study were transplanted with more than 2.0 × 108 cells/kg (Figure 3b). Thus, we could not assess the relationship for patients transplanted with less than 2.0 × 108 cells/kg.

Figure 3

Relation of recipient telomere length, donor age, and the number of infused cells in allogeneic HSCT. The following scattergrams with regression lines are presented. (a) Donor age vs ΔTelallo (= recipient PBMCs–donor PBMCs). (b) The number of infused cells vs ΔTelallo.

Relationship between telomere length and hematological parameters after HSCT

As the telomeres of HSCT recipients were shortened in all cases, we studied the relationship between telomere length and hematological parameters at 1 year after HSCT. There were weak correlations for platelets of autologous HSCT patients (Figure 4a), and in red blood cells and white blood cells of allogeneic HSCT patients (Figure 4b). Three patients (cases 19, 20 and 74) who had markedly shortened telomeres in PBMCs showed impaired recovery of hematological parameters at 1 year after HSCT. The ΔTel of these patients was much greater than those of other recipients (Tables 1 and 2).

Figure 4

Relationship between telomere length and hematological parameters 1 year after autologous and allogeneic HSCTs. The following scattergrams with regression lines are presented. (a) Autologous HSCT patients; (b) allogeneic HSCT patients.


We previously demonstrated that there were no marked differences in telomere length among normal T cells, B cells, monocytes, and polymorphonuclear leukocytes.16 However, recent precise studies demonstrated a significant difference between T cells and neutrophils. Moreover, memory T cells are shown to have shorter telomeres than those of naive T cells.1718 Kronenwett et al19 reported that the mean telomere length of CD34+ cells was about 0.7 kb longer than those of PBMCs. As hematopoietic reconstitution after HSCT involves all series of hematopoietic progenitors and their descendants, we considered that monitoring the telomere length of PBMCs of each patient along with transplantation will reflect the dynamics of the hematopoietic recovery process as a whole.

Previous studies demonstrated that telomeres of recipients are shortened by up to 1.5 kb following HSCT, resulting in the pronounced ‘aging’ of transplanted cells.91011 These findings suggest reconsideration of whether the capacity for complete self-renewal of hematopoietic stem cells lasts for years.12 Wynn et al17 described that this marked telomere shortening represents increased proliferative stress at the level of committed progenitor cells. These studies of telomere length after HSCT focused on younger donors and recipients. In this study, we investigated HSCT-recipients, including patients and donors older than 40 years. The telomere length of HSCT-recipients ranged from 6.8 kb to 12.0 kb, and telomeres of adult patients were shorter than those of children (Tables 1 and 2). The slope of the regression line in autologous HSCT patients was steep compared with that of healthy volunteers (Figure 1a). We have no definite explanation for this phenomenon, but elderly HSCT patients might have shorter telomeres at the time of HSC harvest. Although ΔTelauto correlated with neither patient age nor period after HSCT, the telomere length of over 40-year-old patients was shorter than that of putative age-matched controls. Taken together these data indicate that autologous HSCT patients might have shorter telomeres before transplantation, and that repetitive cytotoxic chemotherapies might account for this telomere shortening.

The significant difference in telomere length between donors and recipients in allogeneic HSCT suggests that allogeneic HSCT requires more numerous divisions of non-self donor stem cells than that of self stem cells in autologous HSCT. The regression line of telomere length after allogeneic HSCT was parallel to that of healthy donors. These results are consistent with Shay12 who suggested that steady-state of hematopoiesis of transplantation is parallel to that pre-transplant.

We found that recipient telomere length and ΔTelallo correlated with donor age rather than with the number of infused cells in allogeneic HSCT. In autologous HSCT, telomere length correlated with patient age, but ΔTelauto did not correlate with patient age. Nataro et al9 described a significant inverse correlation between ΔTelallo and the number of nucleated bone marrow cells infused, but no correlation between ΔTelallo and donor age. This contradictory result may be related to the number of infused cells. In our HSCT protocol, all patients recieved more than 2.0 × 108 cells/kg, while in the Nataro et al study some patients were transplanted with less than 2.0 × 108 cells/kg. The number of HSCs may not be a critical factor for telomere shortening in a case of HSCT with a sufficient number of HSCs.

Mavroudis et al20 demonstrated that patients receiving more than 2 × 106 CD34+ cells/kg in allogeneic HSCT showed significantly earlier recovery of monocytes and a trend toward earlier recovery of lymphocytes. In our study, there was not a strong correlation between telomere length and recovery of hematological parameters in HSCT. However, the patients who had extremely shortened telomeres showed poor recovery of hematological parameters at 1 year after HSCT (Figure 4). Marked shortening of telomere length (6.8 kb) following HSCT may indicate a history of numerous cell divisions of hematopoietic stem cells regardless of impaired hematological state. Hematological parameters of these patients recovered slowly over several years. With regard to telomere length and hematological state, Ball et al15 reported that there was no significant difference in telomere length between recovered and active aplastic anemia patients but a significant correlation between telomere shortening and disease duration. Excessive shortening of telomeres might induce telomeric fusion, leading to the chromosomal rearrangements which are frequently observed in neoplasms. Three of five aplastic anemia patients with excessive telomere loss had acquired an abnormal karyotype,15 and two types of leukemia with telomeric associations have been reported.2122 Moreover, additional chromosome aberrations were found in about 30% autologous HSCT patients,2324 and leukemia rarely occurs in donor-derived cells after allogeneic HSCT.252627

In conclusion, our findings suggest that HSCT for children is safe enough in terms of telomere length, but elderly patients in autologous transplantation or in allogeneic HSCT from elderly donors may develop extremely shortened telomeres. The patients who have less than 7.0 kb telomeres should be observed carefully for their long-term hematopoiesis and the occurrence of hematopoietic disorders.


  1. 1

    Blackburn EH . Structure and function of telomeres Nature 1991 350: 569–573

  2. 2

    Zaikan VA . Telomeres: beginning to understand the end Science 1995 270: 1601–1607

  3. 3

    Harley CB, Kim NW, Prowse KR et al. Telomerase, cell immortality, and cancer Cold Spring Harb Symp Quant Biol 1994 59: 307–315

  4. 4

    Lansdorp PM . Telomere length and proliferation potential of hematopoietic stem cells J Cell Sci 1995 108: 1–6

  5. 5

    Harly CB . Telomere loss: mitotic clock or genetic time bomb? Mutat Res 1991 265: 271–282

  6. 6

    Vaziri H, Schächter F, Uchida I et al. Loss of telomeric DNA during aging of normal and trisomy 21 human lymphocytes Am J Hum Genet 1993 52: 661–667

  7. 7

    Vaziri H, Dragowska W, Allsopp RC et al. Evidence for a mitotic clock in human hematopoietic stem cells: loss of telomeric DNA with age Proc Natl Acad Sci USA 1994 91: 9857–9860

  8. 8

    Hastie ND, Dempster M, Dunlop MG et al. Telomere reduction in human colorectal carcinoma and with ageing Nature 1990 346: 866–868

  9. 9

    Nataro R, Cimmino A, Tabanni D et al. In vivo telomere dynamics of human hematopoietic stem cells Proc Natl Acad Sci USA 1998 94: 13782–13785

  10. 10

    Wynn RF, Cross MA, Hatton C et al. Accelerated telomere shortening in young recipients of allogeneic bone marrow transplants Lancet 1998 351: 178–181

  11. 11

    Akiyama M, Hoshi Y, Sakurai S et al. Changes of telomere length in children after hematopoietic stem cell transplantation Bone Marrow Transplant 1998 21: 167–171

  12. 12

    Shay JW . Accelerated telomere shortening in bone marrow recipients Lancet 1998 351: 153–154

  13. 13

    Lee JJ, Kook H, Chung IJ et al. Telomere length in patients undergoing hematopoietic stem cell transplantation Bone Marrow Transplant 1999 24: 411–415

  14. 14

    Frenck Jr RW, Blackburn EH, Shannon KM . The rate of telomere sequence loss in human leukocytes varies with age Proc Natl Acad Sci USA 1998 95: 5607–5610

  15. 15

    Ball SE, Gibson FM, Rizzo S et al. Progressive telomere shortening in aplastic anemia Blood 1998 91: 3582–3592

  16. 16

    Yamada O, Oshimi K, Motoji T, Mizoguchi H . Telomeric DNA in normal and leukemic blood cells J Clin Invest 1995 95: 1117–1123

  17. 17

    Wynn R, Thornley I, Freedman M, Saunders EF . Telomere shortening in leukocyte subsets of long-term survivors of allogeneic bone marrow transplantation Br J Haematol 1999 105: 997–1001

  18. 18

    Weng NP, Levine BL, June CH, Hodes RJ . Human naive and memory T lymphocytes differ in telomeric length and replicative potential Proc Natl Acad Sci USA 1995 92: 11091–11094

  19. 19

    Kronenwett R, Murea S, Haas R . Telomere length of blood-derived mononuclear cells from cancer patients during G-CSF-enriched marrow recovery Bone Marrow Transplant 1996 18: (Suppl. 1) 10–14

  20. 20

    Mavroudis D, Read E, Cottler-Fox M et al. CD34+ cell dose predicts survival, post-transplant morbidity, and rate of hematologic recovery after allogeneic marrow transplants for hematologic malignancies Blood 1996 88: 3223–3229

  21. 21

    Fitzgerald PH, Morris CM . Telomeric association of chromosomes in B-cell lymphoid leukemia Hum Genet 1984 67: 385–390

  22. 22

    Morgan R, Jarzabek V, Jaffe JP et al. Telomeric fusion in pre-T-cell acute lymphoblastic leukemia Hum Genet 1986 73: 260–263

  23. 23

    Perot C, van den Akker J, Laporte JP et al. Multiple chromosome abnormalities in patients with acute leukemia after autologous bone marrow transplantation using total body irradiation and marrow purged with mafosfamide Leukemia 1993 10: 509–515

  24. 24

    Testoni N, Martinelli G, Zaccaria A et al. Detection of occasional and clonal chromosome aberration in patients with acute non-lymphocytic leukemia after autologous bone marrow transplantation Bone Marrow Transplant 1996 18: 1141–1145

  25. 25

    Giralt SA, Champlin RE . Leukemia relapse after allogeneic bone marrow transplantation: a review Blood 1994 84: 3603–3612

  26. 26

    Elfenbein G, Brogaonkar D, Bias W et al. Cytogenetic evidence for recurrence of acute myelogenous leukemia after allogeneic bone marrow transplantation in donor hematopoietic cells Blood 1987 52: 627–636

  27. 27

    Newburger P, Latt S, Pesando J et al. Leukemia relapse in donor cells after allogeneic bone marrow transplantation New Engl J Med 1981 304: 712–714

Download references


We thank Yoshikatsu Eto (Jikei University School of Medicine) for helpful advice and critical reading of the manuscript. This work was supported in part by the special Coordination Fund for ‘Research for the Future Program’ from the Science and Technology Agency and by a grant in aid from ‘High-Tech Research Center (Institute of DNA Medicine)’ from the Ministry of Education of Japan.

Author information

Correspondence to M Akiyama.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Akiyama, M., Asai, O., Kuraishi, Y. et al. Shortening of telomeres in recipients of both autologous and allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 25, 441–447 (2000) doi:10.1038/sj.bmt.1702144

Download citation


  • hematopoietic stem cell transplantation
  • telomeres
  • elderly recipients and donors

Further reading