Stem Cell Aging

WE summarize here the fruitful discussion that occurred on a wide variety of topics during the session on stem cell aging. Do stem cells age; and if they do, how can these changes be reliably identified? Most attendees were of the view that a decline in the function of tissue-specific somatic stem cells is likely to play an important role in human aging. Results from several mammalian systems were discussed including stem cells of the adult bone marrow, brain, intestine, and muscle. Also, there was a brief discussion of what should constitute a stem cell for the purposes of this discussion, for example, cells like pancreatic β cells, muscle stem cells, and memory T cells appear to share some important features with more classically defined stem cells (eg, an ability to self-renew in adult organisms), but which lack the capacity for broad multilineage differentiation thought to be the sine qua non of a “true” stem cell. Although some aspects of mammalian aging may be shared among such long-term, self-renewing cells with aging, a consensus view emerged that the mechanism and functional consequences of aging are likely to differ markedly by tissue and cell type. For example, transplantable and cell-intrinsic defects and hematopoietic stem cell (HSC) function have been identified by several investigators (1,2), and these defects seem to be increased by stimuli that attenuate DNA repair. In contrast, work of Conboy and colleagues (3) suggests that significant aspects of muscle stem cell aging are cell extrinsic and rather reflect changes in the stem cell milieu (or “niche”; see following). Potential causes of cell-intrinsic aging were discussed. The role of tumor suppressor induction (p53 (4) and p16INK4a (5,6)) in response to cumulative DNA damage, telomere(definition) dysfunction, and other stresses were considered important. These alterations may in turn reflect changes in stem cell chromatin structure, for example, the role of Polycomb group proteins in regulating the INK4A/ARF locus was discussed as a potential contributor to stem cell aging (7). Along these lines, several of the discussants indicated that studies of epigenetic changes, presumably using both candidate approaches and epigenome-wide genomic approaches, should be priorities for future study (see also article by Kahn & Fraga, [8]). Also, an enhanced understanding of stem cell aging was identified as a critical need to provide reliable biomarkers of the process. Potentially useful biomarkers might be changes in chromatin status, changes in gene expression, or changes in measurable macromolecules (eg, telomeres, DNA mutations, protein modifications). What are the roles of stem cell aging (cell intrinsic) and niche aging (cell extrinsic) in the decline of tissue homeostasis in aging and age-dependent risk of disease? The relative contributions of niche versus cell-intrinsic aging were discussed. Several discussants expressed the view that this question may be tissue specific and also that there may be contributions of both types of effects within a given tissue with aging. For example, the work of Ju and associates was (9) discussed showing that animals harboring dysfunctional telomeres have intrinsic functional defects of the HSC as well as extrinsic defects in the bone marrow niche. As changes to the niche (eg, deficiency of some factor that enhances stem cell function) might be particularly tractable to therapy, understanding which features of stem cell functional decline are niche dependent was considered a high-priority question. Several participants wondered if there could be small-molecule approaches to revert niche aging (eg, replacement of missing hormones or inhibitors to block signaling induced by age-promoting ligands in the milieu). Likewise, this discussion led to interest in the topic of whether heretofore unrecognized cell-intrinsic aging could be hampering ex vivo efforts at organogenesis and tissue engineering. What are good approaches to study tissue homeostasis with aging in the absence of injury or regeneration? A particular frustration of discussants was the lack of consensus on validated models for the role of stem cells in aging. As mentioned, the mechanisms of aging may differ among somatic stem cell fractions (eg, intestine vs muscle vs HSC), and therefore, there was concern as to how generalizable results from one of these systems would be to stem cell aging as a whole. In particular, several discussants worried that the experimental tractability of the HSC system has placed undue emphasis on results from this model, whereas the mechanism of stem cell aging of the gut or skin may be very different. Thus, new models may be needed to address such questions as whether the cell-intrinsic aging of constantly proliferating stem cells (eg, gut) is different from the aging of stem cells that are largely quiescent (eg, the HSC) or activated only in response to stimuli (eg, muscle satellite cells). Moreover, concern was voiced in two additional areas regarding model systems: the reliance on “progeroid” mice and humans to inform more general conclusions about mammalian aging, and the frequent need to injure or damage tissues (eg, using transplantation, ionizing radiation, damaging toxins) for regenerative studies. Some discussants voiced the belief that results from such systems may be systematically biased in nonobvious ways. Approaches to study stem cell function with aging in the absence of reliance on germline genetic mutations that bias toward “premature aging” or severe disruptions of normal tissue homeostasis were discussed. It was pointed out that recent advances in mouse genetics using Cre-lox systems and other means of somatic recombination obviate some of the need for globally damaging events and also allow for lineage tracing and fate mapping in the absence of injury and regeneration models. Additionally, induced pluripotency–based systems may provide new human and murine systems for these lines of research (see