A mismatched life in the stem cell pool

In this issue of Blood, Kenyon and colleagues demonstrate that age-related MLH1 hypermethylation leads to DNA mismatch repair defects and microsatellite instability in hematopoietic stem and progenitor cells.¹ 

The relative risks for leukemia, myelodysplastic syndrome, and some lymphoid neoplasms increase with age and it has been proposed that this is at least in part due to acquired genomic instability.²  Hematopoietic stem cells (HSCs) and hematopoietic progenitor cells (HPCs) are constantly exposed to proliferative and environmental stresses that lead to DNA damage and many DNA damage-response systems work in concert to suppress genomic instability. One of these is the highly conserved DNA mismatch repair pathway (MMR).³  MMR corrects chemically-induced base adducts, base mispairing, and slippageevents in single (eg, AAAAAAAA) and double (eg, ATATATAT) nucleotide runs caused by the replicative DNA polymerases. Gains and deletions in nucleotide repeat regions of genomic DNA, referred to as microsatellite instability (MSI), correlate directly with the loss of MMR. MMR is controlled by two central proteins, MLH1 and MSH2, that form heterodimers with other MMR proteins to recognize and orchestrate removal of specific types of DNA damage. Several lines of evidence demonstrate a link between MMR and hematologic disease: homozygous loss of MMR genes correlates with childhood lymphoma and leukemia,⁴  MMR-deficient mice are predisposed to lymphomas,³  and a significant proportion of myelodysplastic and leukemic patients display acquired MMR deficiencies.^5–7  Therefore, because MMR deficiency induces hematologic disease, and because these diseases increase with age and display the hallmark of MMR deficiency, it was reasonable to propose that MMR defects might increase with age in the hematopoietic stem cell pool. Although age-related increases in MSI are found in different cell types, including T lymphocytes,⁸  heretofore the role of MMR-mediated genomic instability in age-related hematopoietic stem cell dysfunction has not been tested.

Kenyon et al examined MSI in CD34⁺ colony forming cell-derived clones (CFCs) from donors ranging in from less than 1 year to 86 years of age.¹  They found that CFCs from donors older than 45 years of age displayed a significantly greater MSI-high frequency than younger donors. Measuring MSI in CFCs was key to this finding as only a fraction of CFCs in individual patients displayed MSI and therefore would not have been detected in whole bone marrow extracts due to MSI signal dilution. Furthermore, MSI correlated with epigenetic loss of MLH1 by hypermethylation (but not MSH2), findings that serve to mechanistically link detected MSI with loss of MMR in HSCs.

These findings provide a model of age-related hematopoietic dysfunction (see figure). Specifically, at birth HSCs and HPCs have fully functional MMR capacities and have few, if any, MSI or point mutations. During the first 45 years of life, a small number of genetic alterations accumulate, but most CFCs retain full MMR capacity. After 45 years, loss of MMR capacity in CFCs is accelerated due to MLH1 hypermethylation resulting in increased genomic instability. It is reasonable to suggest that this accumulation of genetic alterations drives loss of HSC function and/or malignant transformation. The MMR-mediated HSC dysfunction model differs from other genomic instability syndromes where affected cells require adaptive mutations to survive DNA damage.⁹  In contrast, loss of MMR-mediated DNA damage surveillance and signaling makes MMR-deficient cells immediately resistant to specific types of DNA damage.³ 

Age-induced MSI may function as an early marker of hematopoietic dysfunction and provides a target for therapeutic intervention. As MMR capacity is lost due to epigenetic regulation of MLH1, strategies to restore MMR function and MLH1 expression could ameliorate age-related hematopoietic defects. An obvious class of drugs would be demethylating agents; however, these drugs affect global methylation patterns, have serious side effects, and the effects of continual treatment are unknown.¹⁰  Another treatment option may be to selectively target cells that are MMR-deficient; however, because a significant proportion of CFCs are MMR-deficient this would most likely deplete stem cell pools and might make things worse. Therefore, strategies are needed to specifically target MLH1 re-expression, which may not only alleviate age-related hematopoietic dysfunction, but other age-related pathologies.