A recent twin study published in Aging Cell claims to have tested the hypothesis that socio-economic status is associated with telomere attrition independent of known risk factors influencing the aging process (Cherkas et al., 2006). An earlier study, which provided evidence that chronic psychological stress is significantly associated with shorter telomere length (Epel et al., 2004), is consistent with the idea that psychosocial factors, while typically related to socio-economic status, could be more important than socio-economic status per se in relation to health and aging (Sapolsky, 2005). While both papers are mentioned in the Cherkas article, data on psychosocial status among the twins were unfortunately not included. Future longitudinal studies are needed to examine the relationship between stressful characteristics of social rank in relation to telomere length and examine the possible molecular mechanism(s) if correlations are indeed confirmed as expected.
What do we currently know about the molecular mechanisms that influence the telomere length in white blood cells at any given age for a particular individual? Several studies (not cited in the paper) suggest that the primary determinants are genetic (Slagboom et al., 1994; Rufer et al., 1999) and that the telomere length in the precursors of all blood cells, the hematopoietic stem cells themselves, declines with age. The most likely explanation for the latter is that telomerase levels in stem cells are insufficient to compensate for the typical as well as the sporadic loss of telomeric DNA occurring with each cell division (Lansdorp, 2005). Two critical components of telomerase, the telomerase reverse transcriptase (encoded by the TERT gene) and telomerase RNA (encoded by the TERC gene), limit the replication potential of stem cells and can lead to bone marrow failure in patients that are haplo-insufficient for either gene (Vulliamy et al., 2004; Yamaguchi et al., 2005).
Because many different stem cells are expected to contribute to blood cell formation in normal individuals, the relationship between leukocyte telomere length and the telomere length in stem cells is expected to be very indirect. Indeed, it seems possible that acute stress or injury to the hematopoietic system could activate dormant stem cells with longer telomeres and produce a paradoxical increase in telomere length. How changes in the activity and composition of the stem cell compartment relate to telomere length in circulating nucleated cells in the short and long term is currently not known. It seems possible that sustained psychological stress might lead to shorter telomeres through changes in telomere maintenance, stem cell dynamics, or both.
Future longitudinal studies of telomere length in human populations should address the nature of the stress, its duration, and possible mechanisms connecting psychological stressors to stem cell and telomere biology. Such future studies should also take care to study effects on different cell types separately. Leukocyte populations are a very heterogeneous mixture of cells including neutrophils, T cells, monocytes, B cells, etc., each of which may respond quite differently to injury and stress. Previous studies have shown that the decline in telomere length with age is much more pronounced in T cells than in granulocytes (Rufer et al., 1999) and the modest effect of socio-economic status on leukocyte telomere length (Cherkas et al., 2006) may have obscured a much more pronounced effect on, for example, T cells. Fortunately, telomere length can be accurately measured in different subpopulations of leukocytes using flow fluorescence in situ hybridization (FISH), a partially automated technique based on in situ hybridization with a fluorescently labeled telomere probe and flow cytometry (Baerlocher & Lansdorp, 2003). Studies that use these methods to evaluate telomere length in well-defined leukocyte subsets will help to determine whether the observed differences among humans represent alterations in overall telomere length, subset proportions, telomere lengths within subsets, or a combination of these possibilities.