Sci. Aging Knowl. Environ., 26 June 2002
Vol. 2002, Issue 25, p. pe11
[DOI: 10.1126/sageke.2002.25.pe11]


Hematopoietic Stem Cells and Aging

Jennifer Fuller

The author is in the Immunology Program at the University of Michigan, Ann Arbor, MI 48109, USA. E-mail: fullerj{at};2002/25/pe11

Key Words: hematopoiesis • hematopoietic • stem cell • progenitor • blood


Hematopoiesis is a dynamic process by which a single cell--called a hematopoietic stem cell (HSC)--has the proliferative potential to give rise to all of the major lineages of blood and immune cells, including T lymphocytes, B lymphocytes/plasma cells, dendritic cells, natural killer cells, monocytes/macrophages, neutrophils, eosinophils, basophils/mast cells, platelets, and erythrocytes (1, 2). Stem cells can undergo both symmetric division, which results in two copies of the parent stem cell, and asymmetric division, which results in self-renewal of the parent stem cell as well as production of a daughter cell that is capable of differentiating along a specific lineage. Although there is no evidence to indicate that neural crest stem cells and embryonic stem cells persist postnatally, there are ample data to suggest that a single HSC can self-renew and persist throughout adulthood (3, 4).

Hematopoiesis is a highly active, intricately regulated process. In adult mice, ~2.4 x 108 red blood cells (RBCs) and 4 x 106 nonlymphoid peripheral blood cells are produced per day (5). In the adult human, ~1012 blood cells are produced per day, and 4 x 1015 are generated throughout an average lifetime (6). These estimates suggest that to maintain basal hematopoiesis throughout the normal lifetime of a human, an individual HSC must undergo approximately 52 divisions. In addition, during stresses such as blood loss or infection, hematopoiesis must be increased to meet the growing demands of the impaired organism. Therefore, under both normal and pathological conditions, hematopoietic progenitor proliferation must be precisely regulated.

This strict regulation requires that HSCs maintain a functional self-replicating potential throughout the organism's lifetime. There is strong evidence that a number of different elements involved in HSC replication change with age, even though these alterations do not appear to result in hematopoietic deficiencies under normal circumstances. In this Perspective, I discuss the changes that occur with age and why, under most normal circumstances, they do not cause obvious pathologies.

Phenotypes of HSCs: Cell Surface Markers

In the most basic terms, HSCs are defined as the progenitor cells that supply all of the blood cells throughout the life of an organism. In 1988, Berenson et al. (7) used an antibody to human CD34, a protein found on the surface of HSCs and endothelial cells, to identify and select cells capable of reconstituting hematopoiesis in a baboon model. Subsequent studies used antibodies to mouse CD34 to isolate HSCs for transplantation studies, and these cells also were shown to be capable of hematopoietic reconstitution in the murine model. However, in both systems, a purified population of HSCs was found to be a heterogeneous mix of cell types. Distinct subpopulations within this mix have been identified by assessing the expression of lineage marker proteins on the surfaces of individual cells. Further investigation of the murine model showed that, although there were progenitors capable of hematopoietic reconstitution in the CD34+ pool, it was the CD34-/lo, Lin-, Sca-1+, c-Kit+ cell population that was primarily responsible for reconstituting the hematopoietic compartment. This discrepancy resulted from the fact that murine CD34+ cells have short-term self-renewal properties (~6 to 8 weeks); thus, in initial studies, these cells had all of the qualities of stem cells. It was later determined, by using additional antibodies (anti-Sca, anti-c-Kit, and anti-Lin), that CD34-/lo was a more primitive cell that had long-term (years) self-renewal properties. Now, murine HSCs are routinely isolated on the basis of Sca-1 and c-Kit expression, as well as the lack of any specific lineage markers [such as Gr-1 (granulocytes), Mac-1 (macrophages), B220 (B cells), CD4 and CD8 (T cells), and TER-119 (erythroid cells)]. Fractionating HSCs on the basis of lineage marker expression has enabled investigators to define three stem cell populations: (i) Lin-, Mac-1-, CD4- cells, which are the most primitive progenitors and are highly enriched for long-term self-renewal potential; (ii) Lin-, Mac-1lo, CD4-, which are slightly more mature but still have short-term self-renewal properties; and (iii) Mac-1lo, CD4lo HSCs, which have a substantial proliferative potential but are no longer capable of self-renewal (5, 8).

HSC aging is associated with gradual functional and phenotypic changes that are similar to the changes demonstrated in cytokine-mobilized stem cell experimental models, where stem cells are induced to move from the marrow into the bloodstream. HSC populations from old mice contain increased numbers of cells in the active (S, G2, and M) phases of the cell cycle (9), in contrast to HSCs from young animals, which are almost all in a quiescent state. In this respect, HSCs from older animals are similar to HSCs from mice treated with 5-fluorouracil (5-FU) and granulocyte colony-stimulating factor (G-CSF). 5-FU is a pyrimidine analog that has been used for decades as an antineoplastic agent, because it selectively kills dividing cells while sparing quiescent cells. Under normal circumstances, hematopoiesis is highly regulated by complex interactions between the marrow elements and HSCs to maintain a steady state. Steady-state hematopoiesis requires adequate production of numerous differentiated blood cells each day. However, normal marrow also must have the capacity to accommodate rapid responses to acute stress (blood loss or infection or, in this case, 5-FU depletion of cells in the cell cycle) followed by the ability to return to normal blood cell production. When there is an increased demand for cells, the previously quiescent HSCs respond by entering the active phases of the cell cycle. Both 5-FU and G-CSF have been used in combination for stem cell clinical studies and research using murine models. Depending on the stem cell property that is being examined, there is a growing body of data that suggests that HSCs recovered from animals that have undergone these treatments function in a manner that is similar to HSCs from aged donors (10, 11).

Age-related changes in the bone marrow microenvironment (stroma) have also been observed (12). Some of the following changes in bone marrow stromal cells might be responsible for triggering a cell cycle response in normally quiescent HSCs: (i) decreased production of the cytokine interleukin-7; (ii) increased production of prostaglandin E2 (PGE2); and (iii) a shift from a T helper cell-type 1 (TH1; participates in cell-mediated immunity) cytokine profile to a TH2 (participates in antibody-dependent cell-mediated cytotoxicity) cytokine profile. It is important to determine whether these changes are merely indicators of age-related dysfunction or if they actually serve a compensatory purpose to counteract age-related disruptions in hematopoiesis and immune responses.

HSC Reconstitution Properties

Purified populations of HSCs from the bone marrow of young mice have been compared with corresponding HSCs from aged mice. Several in vivo studies have demonstrated that individual HSCs, regardless of donor age, possess the ability to reconstitute all cell types found in the blood (3, 12). However, other studies reveal a number of significant differences between the HSC pools from young versus old mice (8, 12). Surprisingly, the frequency of cells that bear surface markers associated with long-term renewal properties was increased three- to fivefold in aged animals. However, despite their cell surface marker phenotype, these older cells show a decrease (75%) in self-renewal and proliferation capacity, as compared to similarly marked cells from younger animals (4, 12). With respect to these findings, it is important to note the distinction between phenotypic characteristics of long-term repopulating HSCs, such as cell surface markers, and demonstrable characteristics, such as self-renewal for the lifetime of the host. The increased numbers of HSCs in older animals can make up for the reduction in the ability of these cells to self-renew and proliferate. This observation explains why, for several years, bone marrow transplantation (BMT) studies revealed no functional differences between cells of young and old origin. HSCs from aged donors are still capable of reconstituting the marrow, particularly if a bulk bone marrow cell population is used. This is a very common approach for murine studies and is the reason why researchers once believed that HSCs do not age. It is only when these HSCs are examined on a clonal level that the self-renewal and proliferation deficiencies become apparent. These analyses typically are performed by growing a set number of bulk bone marrow cells in a suspension culture so that colonies from individual HSCs can be identified and counted. The number of functional HSCs (that is, those that generate colonies) is about the same for old- versus young-derived cells. Analyses of the surface markers on these cells have shown that old-derived bone marrow contains a higher proportion of cells that are identified as HSCs; but clearly, most of these HSCs are not functioning properly if they cannot proliferate in culture.

Cell Cycle Phases of HSCs

Examination of HSCs from young and middle-aged (less than 15 months of age) mice revealed that almost all of these cells are in the G0/G1 (quiescent) phase of the cell cycle. This means that at any one time under normal circumstances, only 3% of HSCs are stimulated to enter the S/G2/M (active) phases of the cell cycle and that, on average, every HSC enters the active cell cycle phases once every 2 months. The observations of HSCs from old (over 20 months of age) mice told a different story. In the progenitor pool from old mice, 18% of these cells had reentered the cell cycle, which means that more than five times as many HSCs had been driven into the cell cycle as compared to those from their younger counterparts (4). It should be noted that in these experiments, none of the mice younger than 15 months exhibited increased numbers of HSCs in the active phases of the cycle, whereas all of the mice older than 20 months of age showed an increase in the number of cycling HSCs.

Adhesion Molecules and HSC Homing

Normal hematopoiesis and stem cell transplantation require that HSCs successfully migrate in and out of the bone marrow compartment. The mechanisms that regulate HSC movement, or "homing," have yet to be characterized, but multiple classes of adhesion molecules are thought to play a central role in this process. This aspect of stem cell homing was studied using a mouse aging model (strain C57BL/J). Researchers examined age-associated changes in the expression of the adhesion molecule integrin on the surface of HSCs and determined whether these changes were associated with altered homing properties. HSCs from aged mice showed significant down-regulation of the {alpha}4 chain of integrin, and this reduced expression was correlated with a marked reduction in stem cell homing to the bone marrow (13). These changes in integrin expression might affect HSC binding to bone marrow stroma and thereby contribute to the observed reduction in homing efficiency.

Replicative Senescence

Replicative senescence is defined by a cell's finite capacity to divide, along with the various genetic and functional changes that are associated with extensive cell division (14). Under normal physiological conditions, even in old animals, the capacity for HSC replication is not exhausted. In other words, these cells do not undergo senescence. Aged animals do not develop anemia or lymphopenia (a deficiency of white blood cells), and the HSC pool continues to function with no apparent age-related reduction. However, a number of studies in mice have shown that under certain experimental conditions, the capacity of HSCs to replicate is limited. For example, individual HSCs marked with a retrovirus have been observed to lose their replication capacity if they are serially transplanted into a succession of mice.

Although healthy older individuals (mice and humans) do not normally show signs of limited HSC replication, the ability to recover from hematological stresses, such as blood loss or infection, does decline with age. These data suggest that aging might be accompanied by a reduction in the pool of functional stem cells--that is, HSCs that have the ability to self-renew and replicate (5). Another possibility is that HSCs in old organisms have a diminished capacity to respond to signals that induce replication (6). Studies from several groups indicate that both of these scenarios are possible (4-6).

Telomeres and Telomerase

In recent years, telomere shortening has been a topic of great interest to researchers who study mechanisms of aging (see "More Than a Sum of Our Cells"). Telomeres are the noncoding DNA sequences found at the ends of eukaryotic chromosomes. Telomeres mediate chromosome/nuclear matrix interactions, protect DNA from enzymatic breakdown, and are critically involved in cell cycle regulation. Because DNA polymerase does not replicate chromosomal ends completely, telomeres can shorten with each round of cell division, and their diminished length has been used as an indicator of cellular senescence. Telomerase is a ribonucleoprotein with reverse transcriptase activity that preserves and restores telomere length and is found primarily in germline cells, cancer cells, and immortalized cell lines.

From infancy through adulthood, there normally is a gradual decrease in the length of telomeres in peripheral blood leukocytes (15). Indeed, Vazari et al. have demonstrated that telomeric DNA shortens during hematopoiesis (16). Stem cell transplantation studies in mice and humans have also shown that telomeres in peripheral blood leukocytes from the recipient are shorter than those from the donor, suggesting that increased HSC replicative activity results in shortened telomeres. This increase in replication occurs because the recipients are treated with chemotherapy and radiation to deplete their own marrow elements (to vacate "niches" required for successful engraftment) and immune cells (to limit rejection) before BMT. Therefore, the transplanted cells must "rescue" the recipient with extensive proliferation. Telomerase is present in variable amounts in HSCs and peripheral blood leukocytes. Telomerase activity is, in general, low in HSCs, with the lowest amounts observed in fetal liver HSCs and the highest amounts detected in bone marrow HSCs (15). These basal levels of telomerase are not sufficient to prevent the shortening of telomeres during normal replication. In HSCs from aged mice, telomere lengths and telomerase activity are diminished, as compared to young mice.


Despite the fact that individual HSCs appear to function normally with advancing age, the studies discussed here provide mounting evidence that measurable molecular and cellular changes accumulate in the HSC population over time. These changes affect virtually every aspect of the HSC population, from a significant increase in the absolute numbers of these cells, to expansion in the number of HSCs in active phases of the cell cycle, to decreased replicative efficiency, to impaired homing abilities, to altered telomere length and telomerase expression.

It is interesting that these age-associated dysfunctions appear to offset each other, resulting in the overall appearance of normal function. Ordinarily, neither humans nor mice exhibit the anemia or lymphopenia that might be associated with age-related HSC dysfunction. However, numerous deficiencies in the immune response in elderly mice and humans have been documented over the past several years, but a causal relationship between HSC aging and immune dysfunction has yet to be determined. It is now clear that phenotypically similar HSCs derived from young versus old animals are qualitatively different. What has yet to be established is whether these qualitatively different HSCs produce progeny immune cells with modified functional characteristics. A greater understanding of the mechanisms that alter HSC function with advancing aging might aid us in the discovery of innovative therapies to enhance the immune response in elderly individuals.

June 26, 2002
  1. S. J. Morrison, I. L. Weissman, The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity 1, 661-673 (1994).[CrossRef][Medline]
  2. M. Osawa, H. Hanada, H. Hamada, H. Nakauchi, Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 273, 242-245 (1996).[Abstract]
  3. S. J. Morrison, A. M. Wandycz, H. D. Hemmati, D. E. Wright, I. L. Weissman, Identification of a lineage of multipotent hematopoietic progenitors. Development 124, 1929-1939 (1997).[Abstract]
  4. S. J. Morrison, A. M. Wandycz, K. Akashi, A. Globerson, I. L. Weissman, The aging of hematopoietic stem cells. Nature Med. 2, 1011-1016 (1996).[CrossRef][Medline]
  5. S. H. Cheshier, S. J. Morrison, X. Liao, I. L. Weissman, In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 96, 3120-3125 (1999).[Abstract/Free Full Text]
  6. R. B. Effros, A. Globerson, Hematopoietic cells and replicative senescence. Exp. Gerontol. 37, 191-196 (2002).[CrossRef][Medline]
  7. R. J. Berenson, R. G. Andrews, W. I. Bensinger, D. Kalamasz, G. Knitter, C. D. Buckner, I. D. Bernstein, Antigen CD34+ marrow cells engraft lethally irradiated baboons. J. Clin. Invest. 81, 951-955 (1988).
  8. D. E. Harrison, R. K. Zhong, C. T. Jordan, I. R. Lemischka, C. M. Astle, Relative to adult marrow, fetal liver repopulates nearly five times more effectively long-term than short-term. Exp. Hematol. 25, 293-297 (1997).[Medline]
  9. G. S. Hodgson, T. R. Bradley, Properties of hematopoietic stem cells surviving 5-fluorouracil treatment: Evidence for a pre-CFU-S cell? Nature 281, 381-382 (1979).[Medline]
  10. D. J. Richel, E. Van der Wall, J. Slaper, E. Van der Schoot, S. Rodenhuis, Peripheral blood stem cell (PBSC) mobilization and transplantation (PSCT) in patients with malignant lymphomas and solid tumors. Int. J. Artif. Organs 16 Suppl 5, 71-75 (1993).
  11. S. J. Morrison, D. E. Wright, I. L. Weissman. Cyclophosphamide/G-CSF induces hematopoietic stem cells to proliferate prior to mobilization. Proc. Natl. Acad. Sci. U.S.A. 94, 1908-1913 (1997).[Abstract/Free Full Text]
  12. A. Globerson, Hematopoietic stem cells and aging. Exp. Gerontol. 34, 137-146 (1999).[CrossRef][Medline]
  13. A. J. Wagers, R. C. Allsopp, I. L. Weissman, Changes in integrin expression are Lin-/lo Thy1.1lo Sca-1+ c-kit+ hematopoietic stem cells following mobilization by cyclophosphamide/granulocyte colony-stimulating factor. Exp. Hematol. 30, 176-185 (2002).[CrossRef][Medline]
  14. L. Hayflick, Aging, longevity, and immortality in vitro. Exp. Gerontol. 27, 363-368 (1992).[CrossRef][Medline]
  15. M. Engelhardt, R. Kumar, J. Albanell, R. Pettengell, W. Han, M. A. Moore, Telomerase regulation, cell cycle, and telomere stability in primitive hematopoietic cells. Blood 90, 182-193 (1997).[Abstract/Free Full Text]
  16. H. Vazari, W. Dragowska, R. C. Allsopp, T. E. Thomas, C. B. Harley, P. M. Lansdorp, Evidence for a mitotic clock in human hematopoietic stem cells: Loss of telomeric DNA with age. Proc. Natl. Acad. Sci. U.S.A. 91, 9857-9860 (1994).[Abstract/Free Full Text]

Science of Aging Knowledge Environment. ISSN 1539-6150