Editorial: T cell memory, bone marrow, and aging: the good news

Abstract

Discussion on the accumulating evidence that bone marrow in old age is not simply the place where immune cells are generated but the where certain memory cells selectively return to provide a set of distinct immune functions during old age.

Immunological memory is the hallmark of the adaptive immune system. The complex mechanisms by which memory T cells are generated are not completely understood, and the relationship between effector and memory lymphocytes is still being debated. Moreover, the processes involved in the maintenance of the memory phenotype and in controlling the size, stability, and adaptation of memory cells to new specificities have yet to be elucidated [1]. These issues are particularly important in understanding the remodeling of the immune system that occurs during aging, as memory T cells are a predominant component of the immune systems of elderly persons [2]. Many of these memory T cells have features of late-stage differentiation/replicative senescence, such as absence of the CD28 costimulatory molecule and shortened telomeres, indicative of an extensive proliferative history [3]. Moreover, high proportions of these late-stage cells have been correlated with deleterious clinical outcomes, such as reduced vaccine responsiveness, bone loss, and some forms of cancer [4].

With the progressive “graying” of the world population, numerous studies have focused on more detailed characterization of the immunological features of aging, but many gaps and controversies still remain. One contributing factor to some of the conflicting data about immunosenescence relates to the fact that many immunological changes that occur in humans are not easily modeled in short-lived experimental laboratory animals. In particular, the immunological imprint of lifelong chronic infection is a unique feature of humans that cannot be mimicked in the controlled laboratory environment where rodents are housed [5]. Even studies in humans are incomplete, as the majority of immunological data about human aging has been derived from research using PB samples, which at any particular point in time, contain <2% of the total body lymphocyte pool. The few studies that did examine BM and PB in humans were performed on samples from young subjects [6, 7]. The paper by Herndler-Brandstetter et al. [8] in this issue of Journal of Leukocyte Biology begins to address this major research gap by providing evidence for important and unique age-related differences in the phenotype and function of memory T cells within the human BM. The new and somewhat unexpected aspect of the Herndler-Brandstetter et al. study [8] is that in contrast to the PB, the BM is a rich source of polyfunctional memory T cells, which may provide an important defense against recurrent infections in the elderly. Interestingly, a similar accumulation within the BM of cancer patients of less-differentiated, more highly proliferative memory CD8 T cells was also documented recently [9].

The study by Herndler-Brandstetter et al. [8] further underscores the importance of investigating multiple immune compartments to develop a more comprehensive assessment of the unique features of human immunosenescence. That said, this initial study also highlights the fact that the story is far from complete, and important questions remain to be addressed (Table 1). For example, most of the data in this study were derived from persons whom the researchers categorized as old but are still relatively young within the broad spectrum of elderly humans. It will be important to expand this type of research to persons who are in the ″old-old″ category (i.e., greater than age 85). Moreover, it will be of great interest to determine whether exceptional longevity is associated with a unique set of changes within the BM memory compartment, as centenarians have generally avoided most of the chronic diseases of aging and moreover, seem to have a genetic variant of the telomerase enzyme [10], which could possibly retard cellular aging within the memory compartment in general. Another relevant issue is that, whereas the authors show that the BM memory CD8 T cells are able to proliferate in short-term culture in response to certain cytokines, more long-term and extensive cell division will invariably, ultimately result in critically short telomeres, DNA damage, and p53-mediated and cell cycle arrest, so that by extremely old age, these BM memory T cells may acquire different phenotypic and functional features. For example, if the memory T cells within the BM eventually reach the end stage of replicative senescence, their cytokine profiles may start favoring enhanced maturation and activity of bone-resorbing osteoclasts, thereby possibly contributing to age-related bone loss [11].

Table 1. Future Research Questions Regarding BM T Cells.

1. Effects of aging

What is the antigen specificity of the memory CD8 T cells in the BM in older persons?

What is the effect of aging on the BM memory T cell repertoire?

How does aging affect the CD4 memory T cells within the BM?

Are the multifunctional features of memory T cells retained even after age 80?

Is extreme longevity associated with unique BM memory T cell characteristics?

What is the effect of frailty on memory T cells in the BM?

How does aging affect memory T cells in the gut, skin, and LNs versus PB?

2. BM T cell homing and egress characteristics

What factors (chemokines, integrins, adhesion molecules, etc.) control homing to the BM?

What regulates the balance between homing to and egress from the BM

How do aging, exercise and stress affect trafficking of T cells into and out of the BM?

Does aging affect the distribution of T cells within different bones?

3. BM local microenvironment

What are the cellular components of the T cell niche within the human BM?

Are CD4 and CD8 T cells localized in the same BM niche during aging?

What mechanisms regulate the number of T cells within the BM?

Do T cells within the BM affect certain aspects of hematopoiesis?

How does aging impact the interaction between T cells and hematopoietic stem cells?

4. BM T cells and disease

Do T cells in BM contribute to pathological bone loss associated with inflammatory states?

Can BM T cell homing and egress be manipulated for treatment of diseases?

How might BM T cell function be enhanced for protection against tumors or viral infections?

The population studied by Herndler-Brandstetter et al. [8] was a convenience sample, based on the availability of remnant bone acquired for, presumably, cosmetic surgery, suggesting that the subjects may have been in a rather high socioeconomic group and were in relatively good health. This raises an important caveat regarding the results, in light of a recent analysis of ∼400 subjects, which documented that socioeconomic status has a dramatic effect on the rate of biological aging. The study demonstrated that persons who are at the lower end of the socioeconomic scale showed more rapid biological aging (based on the rate of telomere loss within PB cells) and had shorter lifespans as compared with persons at the higher end of the economic spectrum [12]. Thus, increased numbers, age spans, and socioeconomic categories of subjects are required to characterize more fully the effect of aging on the BM memory T cell compartment. It will also be of interest to examine memory T cells in BM from sites other than the iliac crest (the source of BM cells in this study), as well as cells within the gastrointestinal tract and LNs, and compare memory cells within these other compartments with cells in the PB. In addition, there will be increasing interest in more general questions regarding BM biology in elderly humans, such as whether age alters T cell/stromal cell interactions within the BM [1, 13] and how the BM memory T cell profiles might correlate with overall health and longevity.

A final aspect of the BM memory T cell pool that merits more careful investigation relates to the antigenic specificity of the cells within this compartment [6]. The repertoire of different antigenic specificities may provide novel insights regarding histories of viral reactivation patterns. A rough calculation, based on estimated prevalence of serum antibody data, suggests that there are many billions of chronic viruses within the earth's total human population and that each of us harbors approximately eight to 12 chronic infections [14]. Based on PB data, one might predict that a large proportion of memory T cells within the BM is directed at CMV and possibly other persistent viruses. Indeed, several studies have begun to address this issue and have, in fact, shown high proportions of CMV- and EBV-specific CD4 and CD8 T cells within the BM compartment, and CMV-specific central memory T cells are even higher than in the PB [15, 16].

In conclusion, the accompanying paper provides intriguing new data about a novel facet of immune function within old persons' BM. The work adds to the accumulating evidence that the BM is not simply the place where immune cells are generated. Indeed, the new study suggests that certain memory cells can selectively return to their place of birth and provide a set of distinct immune functions that may possibly enhance immunity during old age.