| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Stem Cell Biology, University of Groningen, Groningen, The Netherlands;
b Department of Hematology, University Hospital Groningen, Groningen, The Netherlands
Key Words. Hematopoietic stem cells • Aging • Transplantation • Genetics
Correspondence: Gerald de Haan, Ph.D University of Groningen, Department of Stem Cell Biology, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands. Telephone: 31-50-363-2722; Fax: 31-50-363-7477; e-mail: g.de.haan{at}med.rug.nl
 |
ABSTRACT |
|---|
 Top Abstract IintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
IINTRODUCTION |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the study reported here, we assessed whether an intrinsic, genetic program exists that leads to stem cell exhaustion in serial transplantation models. To this end, we estimated the maximal number of cobblestone area–forming cell (CAFC) d35 population doublings in two genetically distinct mouse strains, C57BL/6 (B6) and DBA/2 (D2) mice, which differ with respect to numerous HSC traits [17–20]. During normal aging, the number of stem cells and their repopulating ability after transplant decreased in D2 mice, whereas these parameters increased in B6 mice [17, 19, 21–27]. Intriguingly, D2 mice have a shorter life span than B6 animals have [28]. The presence of a mitotic clock operating in proliferating cells and potentially limiting the lifespan of cells in culture has been a topic of much debate ever since it was first postulated by Hayflick [29, 30]. This applies, in particular, to the relevance of such a putative clock in limiting organismal lifespan. Since adult somatic stem cells have a key role in replenishing tissues, it is interesting to speculate that conservation of stem cell functioning during normal aging extends lifespan.
In this study we used two distinct transplantation models, unfractionated bone marrow cells or highly purified Lin–Sca-1+c-kit+ (LSK) cells, to estimate the maximal number of CAFC d35 population doublings following serial transplantation. Using unfractionated BM, the number of transplanted stem cells can only be assessed retrospectively, increasing the likelihood that HSC dose varies from one transplant to the other, potentially resulting in stem cell dilution. This problem is partly controlled by using highly purified stem cells, although it has been shown (and was confirmed in the current study) that functional activity per stem cell, based on phenotype (Lin–Sca-1+c-kit+) declines in serial transplantations. Finally, we tested whether BM cells isolated from a young donor were able to engraft in recipients that were previously reconstituted with serially transplanted BM cells or to engraft in nonconditioned older recipients.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Serial Transplantations
Bone marrow cells were isolated from both femora. Single-cell suspensions were obtained by flushing each femur three times, and the total number of nucleated cells was assessed using a Coulter Counter (Coulter Electronics, Dunstable, England). Prior to purification of HSCs, a standard NH4Cl erythrocyte lysis was performed. Subsequently, cells were incubated with 5% normal rat serum for 15 minutes at 4°C, after which cells were stained with biotinylated lineage-specific antibodies (anti-B220, anti-Gr-1, anti-Mac-1, anti-TER-119, and anti-CD3e), fluorescein isothiocyanate (FITC)–anti-Sca-1, and Allophycocyanin (APC)–anti-c-kit (all antibodies from Pharmingen, San Diego, CA; http://www.bdbiosciences.com) for 40 minutes at 4°C. Cells were subsequently incubated with Streptavidin-PE (Pharmingen) for 40 minutes at 4°C and strained through a 35-µm cell strainer (Becton, Dickinson, Bedford, MA; http://www.bd.com), after which cells were purified using a MoFlo flow cytometer (Dako Cytomation, Fort Collins, CO; http://www.dakocytomation.com). The clearly distinct Sca-1+c-kit+ population among the 5% of the BM cells showing the least phycoerythrin (PE) fluorescence intensity was selected. Sorting gates were determined based on normal BM cells and were not changed when serially transplanted samples were analyzed. Recipients were lethally irradiated (9.5 Gy) using an IBL 637 Cesium-137 
source (CIS Biointernational, Gifsur-Yvette, France, http://www.cisbiointernational.fr/) 24 hours prior to transplantation of either 4 x 106 unfractionated BM cells or 1,500 sorted Lin–Sca-1+c-kit+ (LSK) cells. Two independent sets of experiments were performed with LSK cells. At 4–6 months after each transplantation, recipients were sacrificed, bone marrow was collected, and the entire procedure was repeated with transplantations in a new group of lethally irradiated recipients. Both transplantation models were initiated with 6–12 recipients. Due to lack of cells and loss of survival of recipients, in selected groups, fewer recipients were used in later transplantations.
CAFC Assay
In vitro limiting dilution-type long-term BM cultures were used to assess clonogenic activity in unfractionated and purified cell fractions. To this end, the CAFC assay was performed, as described previously [17, 34].
Where as committed progenitors form colonies 7–14 days after culture initiation, the most primitive hematopoietic cell subsets appear late (i.e., after 35 days) in culture [35].
Determination of Donor Cell Contribution after Serial Transplant
To verify donor-derived hematopoiesis, the presence of the Y-chromosome in individual progenitor cell colonies was assessed at various time points. To this end, methylcellulose colony-forming unit granulocyte-macrophage (CFU-GM) cultures were initiated. After 7 days of culture, DNA was isolated from individual colonies, as described previously [19]. To assess the presence of the Y-chromosome, a 20-µ1 polymerase chain reaction (PCR) (10 minutes at 95°C; 50 cycles of 30 seconds at 94°C, 30 seconds at 60°C, and 30 seconds at 72°C; 10 minutes at 72°C) was performed using a Perk-inElmer GeneAmp PCR system 9700 (PerkinElmer Corp., Norwalk, CT; http://www.perkinelmer.com) and Y-chromosome–specific primers (Isogen Bioscience B.V., Maarssen, The Netherlands; http://www.isogen-lifescience.com); YMT2/B1 5'-CTG GAG CTC TAC AGT GAT GA-3' and YMT2/B2 5'-CAG TTA CCA ATC AAC ACA TCA C-3', amplifying a 342-bp fragment. PCR products were analyzed using a 1% ethidium bromide-stained agarose gel.
Competitive Repopulating Assay
To determine in vivo reconstituting potential, peripheral white blood cell numbers were counted using a Coulter Counter (Coulter Electronics), and, after serial transplantation was performed, hematocrits were obtained.
Competitive repopulating assays were carried out in two distinct settings. First, 4 x 106 B6 BM cells that had been serially transplanted five times were mixed with an equal number of BM cells that were freshly isolated from young eGFP+ transgenic B6 donors [32, 33], and both cell populations were co-transplanted in lethally irradiated B6 recipients. At various timepoints after transplantation, the levels of eGFP+ leukocytes in the peripheral blood was quantified using fluorescence-activated cell sorter (FACS) analysis (FACS Calibur; Becton, Dickinson, Palo Alto, CA; http://www.bd.com). Second, to address the issue whether the engraftment defect of serially transplanted BM was due to a homing defect, lethally irradiated B6 recipients were reconstituted with 5x serially transplanted BM cells alone, and after 3 months the freshly isolated BM cells from young eGFP+ B6 donors were infused, without additional conditioning.
Bone Marrow Transplantations in Nonirradiated Young and Aged Recipients
Four to five million unfractionated BM cells isolated from 6- to 8-week-old female eGFP+ donors were infused intravenously in nonirradiated, normal, healthy, young (< 6 months old) or aged (>16 months old) wild-type B6 recipients. Transplantations were repeated every 2 months. Prior to each transplant, the presence of eGFP+ leukocytes in the peripheral blood was assessed by flowcytometry.
Survival of old, transplanted recipients was compared with a cohort of aged-matched non-transplanted animals, housed under the same conditions.
Quantification of Expansion and the Number of Stem Cell Population Doublings
The total number of CAFC d35 present in the whole animal at the moment of sacrifice was calculated from the number of CAFC d35 present in one femur, assuming that one femur represents 6% of the total marrow [36, 37]. The fold expansion of the CAFC d35 was calculated by assessing the total number of CAFC d35 present per animal and dividing this number by the number of infused CAFC d35 some 4–6 months earlier. The actual number of CAFC d35 population doublings was calculated by the algorithm: (2n = fold expansion), in which n = the number of population doublings [38].
Statistical Analysis
During serial transplantation, BM cells of recipients were pooled and CAFC cultures were initiated. The entire set of serial transplantations using purified HSCs was repeated twice. Values of the two independent experiments of LSK transplantations were averaged, and the SEM was calculated. Kaplan-Meier survival analysis was done of aged mice that received repetitive eGFP+ transplantations and of the non-treated controls. Differences in survival were tested for significance using a log-rank test. Statistical analysis was performed using the SPSS statistical package. The Student’s t-test was used to assess significant differences in expansion of D2 and B6 unfractionated BM cell and LSK transplantations.
|
RESULTS |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Survival was better in recipients transplanted with unfractionated BM cells. The transplantations of highly purified HSCs ended after the fourth step; due to survival of too few recipients, isolation of sufficient numbers of LSK cells for further transplantation could not be achieved.
|
).
In recipients transplanted with unfractionated BM cells, we did assess the percentage of original male donor cells by detection of Y-chromosome–specific sequences in individually isolated CFU-GM colonies after the third and sixth transplantation. After the third transplantation, >95% of the progenitors were of donor (male) origin in both D2 and B6 mice (Fig. 1A). In contrast, after the sixth transplantation, only few colonies (<15%) colonies were donor-derived (Fig. 1B), indicating exhaustion of the original transplanted stem cells and recovery of heavily irradiated host stem cells.
|
As no substantial part of the hematopoietic system was derived from the original male donor after the sixth transplantation, the ones using unfractionated BM were ended after the fifth transplant.
Stem Cell Numbers after Serial Transplantations
When 1,500 purified HSCs were transplanted, a continuous decrease of CAFC d35 frequencies in D2 mice was observed (Fig. 2A). In contrast, frequencies in B6 mice increased at first, and only after three transplantations was a severe decline noted. Using unfractionated BM, CAFC d35 frequencies decreased in B6 mice, but, again, more dramatically in D2 animals after the first three transplantations (Fig. 2B). Hereafter, the frequency continued to decline for D2 animals but stabilized in B6 mice. The decline of CAFC d35 frequency in combination with lack of donor contribution clearly indicates functional exhaustion of stem cells.
|
). No difference was seen between the frequency of LSK cells in D2 and B6. The decline of the percentage of LSK cells in the BM of the recipients resulted in a gradual reduction of the total LSK pool size per animal (data not shown). In addition, in B6 mice a minor decrease was observed for the clonogenic CAFC d35 activity of the purified cells, until a sudden drop after the fourth transplant (Fig. 2D). This suggests severe functional impairment of these serially transplanted HSCs. In D2 mice, the clonogenic activity returned to normal levels after a small decline after the first and second transplant.
Competitive In Vivo Assays
Although it has been well documented that during normal, unperturbed hematopoiesis there is a strong overlap between LSK FACS phenotype, in vitro functional CAFC activity, and in vivo long-term repopulating ability, this correlation is not always evident in specific experimental conditions in which the system is perturbed, such as serial transplantations [15]. Therefore, to confirm functional stem cell decline measured by FACS (Fig. 2C) or CAFC potential (Fig. 2A–B, D), we mixed 5x serially transplanted unfractionated B6 BM cells with an equal number of freshly isolated eGFP+ young BM cells, and co-transplanted both cell populations in lethally irradiated B6 recipients. Although 5x serially transplanted BM, which consist of a mixture of original donor cells and irradiated recipients, possessed substantial numbers of CAFC d35 (Fig. 2B), there was no evidence of any in vivo stem cell activity when transplanted with similar numbers of CAFC d35 derived from young competitors (Fig. 3A). However, when 5x serially transplanted BM was transplanted without competitors (i.e., a 6th transplant), all recipients survived, clearly documenting that remaining stem cells were present and engrafted (Fig. 3B). To further illustrate that after serial transplantations stem cells have acquired a dramatic competitive defect when compared with freshly isolated BM that is not due to a difference in homing, we transplanted, without further conditioning, eGFP+ unfractionated BM cells in animals that had received a transplant of 5x serially transplanted BM cells 3 months earlier. Freshly isolated BM cells engrafted very rapidly in these recipients, and hematopoiesis converted fully to an eGFP+ genotype (Fig. 3B), which confirms a competitive disadvantage of serially transplanted cells.
|
) that the FACS phenotype is not very appropriate, as functional activity of phenotypically identical cells is highly affected by serial transplantations [15]. In vivo competitive repopulation assays have been used frequently and are very informative. However, in our current study we show that although serially transplanted BM cells may have no competitive repopulating potential, they may still contain stem cells, as they repopulate recipients when transplanted alone (Fig. 3A–B), and stem cell activity in these cells was detected by the CAFC assay (Fig. 2B). Therefore, although we do not claim that the CAFC assay is superior to other assay models in all circumstances, it was deemed valuable in the context of the current experiments.
The actual expansion of CAFC d35 after each transplant was calculated by dividing the total number of CAFC d35 present in the bone marrow in the entire mouse 4–6 months after transplantation, by the number of CAFC d35 infused (Fig. 4). Expansion of D2 CAFC d35 was stable (approximately 30-fold) during each serial transplantation of 1,500 LSK cells (Fig. 4A). Expansion of B6 CAFC d35 started at a much higher value (approximately 700-fold) but decreased rapidly with each transplant (Fig. 4A). During serial transplantation of unfractionated BM, the expansion of D2 CAFC d35 was again stable (approximately 15-fold), similar to transplanting purified LSK cells (Fig. 4B). Expansion of B6 CAFC d35 increased during the first three transplantations from 23-fold to 114-fold but dropped after the third transplant to ultimately 81-fold during the fifth transplant. The total cumulative expansion, which is the product of the expansion after each transplant, was dramatically higher (>1,000-fold) in B6 mice than in D2 mice, but similar whether purified LSK cells or unfractionated BM was used. (Fig. 4C–D).
|
). Remarkably, the same maximum number of population doublings was reached when either highly purified HSC or unfractionated BM was used for serial transplantation. However, whereas the maximum number of population doublings for B6 was approximately 30, D2 stem cells could undergo only approximately 20 population doublings. The strain differences were already apparent after three serial transplantations of unfractionated BM cells.
|
). Repeated transplantations resulted in further increases of donor-type engraftment (Fig. 6B), which was multilineage (Gr-1+, Mac1+, B220+, CD3+; data not shown).
|
,p = .108). |
DISCUSSION |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Clearly, these values are an approximate estimate of stem cell self-renewal capacity. First, we have shown that recipient cells contribute to hematopoiesis in the final serial transplantations, which interferes with our calculation. Second, our estimate is assay-dependent and holds true when CAFC d35 are used as endpoints. This is illustrated by lack of engraftment in competitive assay while CAFC d35 activity was still observed. Finally, our estimate only applies to the entire population of cells, and we cannot rule out that individual, single HSCs can exceed this threshold. Importantly, the first two arguments imply that our calculations are an overestimate of the true self-renewal potential of the HSC population. It is unlikely that these limits of the amplification potential are reached during normal aging. To put this into perspective, in a study where B6 stem cells were labeled in vivo by BrdU, it has been shown that it takes ~2 months before >99% of all HSCs have undergone a cell division [40].This would suggest that during their entire lifespan (mean lifespan ~ 28 months) HSCs undergo ~14 divisions.
Nonetheless, our data clearly reveal that young BM cells provide multilineage engraftment in aged (>16 months) recipients, as well as in recipients of five time serially transplanted BM cells. As this was not observed in young recipients, we believe that these engraftment levels indicate a functional defect of endogenous stem cells in old mice and in serially transplanted BM cells. However, this stem cell defect is very mild in aged recipients, and, consequently, the levels of engraftment are modest. Therefore, we were not surprised that repeated cell transfer did not extend lifespan, although the trend that we observed is intriguing. Much larger groups of mice will be needed to rigorously test the lifespan-extending potential of HSC transplantations. In addition, B6 mice may not be the most optimal model to test this, as these mice are relatively age-resistant. Interestingly, it has recently been shown that caloric restriction, currently the only effective way to extend lifespan in rodents, preserves stem cell function in old BALB/c mice [41].
A number of potential explanations for stem cell exhaustion may be coined, among which the hypothesis of telomere loss during cell division is the most frequently supported one. Decreased telomere length of stem cell chromosomes after each transplant is a potential mechanism that could limit the maximum number of self-renewal divisions. It has been postulated that loss of telomeric DNA with age might function as a mitotic clock during normal aging [42, 43], as well as during serial transplantation of HSCs [44]. In addition, an association has been made between the self-renewal potential of HSCs and telomerase activity [45]. Importantly, stem cells isolated from telomerase (mTERT)–deficient mice can only be serially transplanted twice. This is accompanied by a rapid decline in telomere shortening in HSCs and reduction of long-term repopulating capacity [39, 46]. However, although HSCs overexpressing mTERT showed maintenance of telomere length after serial transplantation, no extended replicative capacity was observed, compared with wild-type mice [47]. Recently, it has been suggested that telomere length as such is less important for the process of aging than specific molecular alterations in telomere structure during aging [48, 49]. The role of telomere shortening and telomerase activity in HSCs of mice is further obscured as mice have longer telomeres than humans have [50]. With respect to our current results, it is unlikely that telomere erosion contributes to the reduced self-renewal ability of D2 HSCs as D2 animals have longer telomeres than B6 mice have [51].
Another process that might affect in vivo stem cell self-renewal is proliferative activity of hematopoietic cells, which is consistent with a previously suggested model that proliferation leads to loss of stem cell quality [52]. The higher cycling activity of D2 progenitor cells might lead to more rapid exhaustion of D2 HSCs [17]. However, although D2 CAFC d35 made fewer self-renewing population doublings, we did not observe faster exhaustion of these cells, so other mechanisms are involved as well. Theoretically, HSCs have the option to either differentiate, self-renew, or die. During steady-state hematopoiesis the self-renewal probability is .5. We speculate that D2 HSCs, during serial transplantation, have a preference to differentiate rather than to self-renew. This would explain that, although there is a lower expansion per transplant and a lower number of maximal population doublings, the white blood cells and hematocrit in D2 mice are stable for a sustained period of time. Therefore, a reduced self-renewal probability (p < .5) to sustain blood cell counts would result in lower expansion. In agreement, it has been shown that peripheral blood cell counts are a poor indicator of marrow stem cell reserve [11, 53]. A decreased preference of D2 HSCs to self-renew would also explain the gradual decrease of stem cells in D2 mice during normal aging, whereas stem cell numbers increase in B6 mice [22, 23, 54]. Moreover, it provides a plausible explanation for the skewing of D2 and B6 leukocyte contribution in D2-B6 chimeras in favor of B6 stem cells [19, 24].
In conclusion, our data clearly establish that maximum self-renewal capacity of HSCs is genetically controlled. The genetic balance between differentiation and stem cell self-renewal may affect stem cell aging and may possibly affect organ lifespan.
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIintroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. H. Cho, H. B. Sieburg, and C. E. Muller-Sieburg A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells Blood, June 15, 2008; 111(12): 5553 - 5561. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. van Os, L. M. Kamminga, A. Ausema, L. V. Bystrykh, D. P. Draijer, K. van Pelt, B. Dontje, and G. de Haan A Limited Role for p21Cip1/Waf1 in Maintaining Normal Hematopoietic Stem Cell Functioning Stem Cells, April 1, 2007; 25(4): 836 - 843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Schatteman, M. Dunnwald, and C. Jiao Biology of bone marrow-derived endothelial cell precursors Am J Physiol Heart Circ Physiol, January 1, 2007; 292(1): H1 - H18. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Xing, M. A. Ryan, D. Daria, K. J. Nattamai, G. Van Zant, L. Wang, Y. Zheng, and H. Geiger Increased hematopoietic stem cell mobilization in aged mice Blood, October 1, 2006; 108(7): 2190 - 2197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Kamminga and G. de Haan Cellular Memory and Hematopoietic Stem Cell Aging Stem Cells, May 1, 2006; 24(5): 1143 - 1149. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. V. Priestley, L. M. Scott, T. Ulyanova, and T. Papayannopoulou Lack of {alpha}4 integrin expression in stem cells restricts competitive function and self-renewal activity Blood, April 1, 2006; 107(7): 2959 - 2967. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Kamminga, L. V. Bystrykh, A. de Boer, S. Houwer, J. Douma, E. Weersing, B. Dontje, and G. de Haan The Polycomb group gene Ezh2 prevents hematopoietic stem cell exhaustion Blood, March 1, 2006; 107(5): 2170 - 2179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Yu, Y. Yuan, H. Shen, and T. Cheng Hematopoietic stem cell exhaustion impacted by p18INK4C and p21Cip1/Waf1 in opposite manners Blood, February 1, 2006; 107(3): 1200 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bhattacharya, D. J. Rossi, D. Bryder, and I. L. Weissman Purified hematopoietic stem cell engraftment of rare niches corrects severe lymphoid deficiencies without host conditioning J. Exp. Med., January 23, 2006; 203(1): 73 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Zipori The Stem State: Plasticity Is Essential, Whereas Self-Renewal and Hierarchy Are Optional Stem Cells, June 1, 2005; 23(6): 719 - 726. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Haider and M. Ashraf Bone marrow stem cell transplantation for cardiac repair Am J Physiol Heart Circ Physiol, June 1, 2005; 288(6): H2557 - H2567. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||
