A (DNA) stitch in time saves nine

Long-lived cells, especially those with the extensive developmental and proliferative potential to produce large numbers of progeny, are typically endowed with robust powers for detecting and repairing DNA damage acquired through the rigors of replication, transcription, and long-term exposure to the environment. Stem cells are perhaps the best example of such long-lived cells since they must replace worn-out progeny throughout a lifetime. Failure of stem cells to maintain DNA fidelity, and hence genomic stability, has several possible outcomes that range from poor to dire. The least onerous is effective recognition of the damage by the cell's surveillance arm and elimination of the cell via effector mechanisms that include apoptosis or the induction of senescence. Both outcomes prevent the participation of, for example, damaged hematopoietic stem cells (HSCs) in blood cell formation and preclude dissemination of genetic abnormalities. A disadvantage of this scenario is that stem cells incapable of effective DNA repair deplete an already small population. If the threshold for activation of the apoptosis or senescence pathways is more stringent, depletion may be accelerated. And if the apoptosis-senescence threshold is more relaxed, stem cell depletion may be slowed but at the expense of potentially retaining stem cells with irreparable DNA damage. The poorest outcome is the loss of a stem cell's capacity to either sense and/or repair DNA damage, an outcome that runs a high risk of genomic instability and tumorigenesis. Thus, a stem cell's powers of surveillance and repair are crucial and may provide an important intersection of the fields of cancer biology and the biology of aging.

A major mechanism by which aging causes decreased organ function is cumulative damage to a cell's macromolecules, notably its DNA (see Johnson et al, Cell. 1999;96:291-302). Since the single largest risk factor for the development of cancer is advancing age, it is tempting to conceptually link the two (see DePinho, Nature. 2000;408:248-254). Moreover, since tumors are often clonal progeny of a genomically unstable founder, stem cells whose normal function is the production of large numbers of progeny are attractive candidates for the source of dysregulation in tumorigenesis. Hematologists are familiar with the concept of aberrant stem cells being the source of hematologic malignancies, and the concept is gaining favor in the study of solid tumors as well (see Al-Hajj et al, Proc Natl Acad Sci U S A. 2003;100:3983-3988).

Reese and colleagues (page 1626) show that in mice null for an important effector of DNA mismatch repair, MutS homolog 2 (MSH2), hematopoietic stem cells possess an enhanced capacity to survive exposure to the methylating agent, temozolomide (TMZ), a compound that causes DNA damage requiring mismatch repair mechanisms. MSH2-null stem cells had a strong competitive survival advantage relative to wild-type stem cells and were able to successfully restore hematopoiesis following the myelosuppressive effects of TMZ. However, the neglected stitch in time carried the heavy price of MSH2^–/– HSCs being unable to contribute to the engraftment of recipients of serial transplants, culminating in stem cell exhaustion, marrow aplasia, and death. Surveillance mechanisms of MSH2-null stem cells apparently detected the accumulating DNA damage in secondary and tertiary transplant recipients, possibly inflicted by the replicative stress of serial engraftment demands. Effector mechanisms may then have either purged or silenced the defective stem cells, thus demonstrating that not even nine belated stitches were able to save the day. One might expect that surveillance and purging would fall short of perfect in the context of a greatly heightened background of genetic damage. Although tumorigenesis (in this scenario it is typically lymphomagenesis) was not detected in these studies, longer-term studies of larger numbers of transplant recipients would likely have revealed the emergence of lymphoma in hosts receiving MSH2^–/– HSC transplants. Reese et al ruled out the possibility that either diminished marrow homing capacity or perturbed cell cycle kinetics accounted for their observations. Moreover, the telomere lengths of blood cells derived from MSH2^–/– and wild-type stem cells remained comparable following transplantation, demonstrating that the loss of functioning MSH2^–/– HSCs was not triggered through this mechanism. Rather, the fact that the authors detected an increase in microsatellite instability in clones derived from Sca-1⁺Kit⁺Lin^– cells of MSH2^–/– mice argues strongly for genomic instability being the cause of stem cell exhaustion. Animal models such as the one employed by Reese et al may aid in further strengthening the emerging links between aging, DNA repair, cancer, and stem cells.