T cell–mediated immune responses to an antigen usually result in the generation of memory T cells specific for that antigen, which may persist for years, even a lifetime. Memory cells provide effective defense against pathogens that are prevalent in the environment and may be repeatedly encountered. Despite the importance of immunologic memory, many fundamental questions about the generation and maintenance of memory cells have still not been answered.
The mechanisms that determine whether an individual anti gen-stimulated T cell will become a short-lived effector cell or enter the long-lived memory cell pool are not fully established, and there are three major competing models. Memory cells may develop from effector cells along a linear pathway, meaning that as effector T cells contract, a small population of memory pre cursor cells remains from which memory cells are generated (Fig. 1A). An alternative model proposes that some naive cells differentiate through multiple independent stages into memory T cells following activation. Additional stimulation may induce some of the memory cells to differentiate into terminal effector T cells that die rather than contribute to durable memory. This model implies unidirectional differentiation (Fig. 1B). Lastly, it has been proposed that effector and memory populations follow divergent differentiation and are two alternative fates of lymphocytes activated by antigen and other stimuli (Fig.1C). The signals that drive the development of memory cells are also not fully understood. These signals may include the strength of TCR stimulation, the level of costimulation, the cytokine environment, and others. No single transcription factor determines whether an antigen-stimulated T cell will become a terminal effector cell or a memory cell; rather, this choice may be controlled by quantitative differences in numerous transcription factors and epigenetic reprogramming.
Fig1. Development of memory T cells. (A) Following activation, T cells develop into effectors, some of which survive as memory cells. (B) Activated T cells differentiate into self-renewing stem cell-like populations that can give rise to memory and effector cells. (C) Whereas memory T cells are maintained, effector cells die by apoptosis after performing their functions. Tcm , central memory T cells; Teff , effector T cells; Tem , effector memory T cells; Tscm , stem cell-like memory T cells.
Properties of Memory T Cells
The defining properties of memory cells are their ability to survive for prolonged periods after an antigen is eliminated and to mount larger and more rapid responses to antigens than do naive cells. Several features of memory cells account for these properties.
• Memory cells express increased levels of antiapoptotic proteins, which may be responsible for their prolonged survival. Whereas naive T cells live for weeks or months and are replaced by mature cells that develop in the thymus, memory T cells may survive for years. Thus, as humans age in an environment in which they are constantly exposed and responding to infectious agents, the proportion of memory cells induced by these microbes compared with naive cells progressively increases. In individuals older than 50 years of age, half or more of circulating T cells may be memory cells. The anti-apoptotic proteins that promote memory cell survival include BCL-2 and BCL-XL , which block apoptosis induced by a deficiency of survival signals. The presence of these proteins allows memory cells to survive even after antigen is eliminated and innate immune responses have subsided, when the stimuli for effector T-cell survival and proliferation are no longer present.
• Memory cells respond more rapidly to antigen stimulation than do naive cells specific for the same antigen. This is one reason why secondary responses to antigen exposure are more rapid than primary responses. A possible explanation for this accelerated response is that the gene loci for cytokines and other effector molecules are fixed in an accessible chromatin state in memory cells, in part because of changes in methylation and acetylation of histones. These epigenetically modified genes are poised to respond rapidly to antigen recognition.
• The number of memory T cells specific for any antigen is greater than the number of naive cells specific for the same antigen. As we discussed earlier, proliferation leads to a large clonal expansion in all adaptive immune responses, and the memory cells that remain from the expanded clone are typically 10- to 100-fold more numerous than the pool of naïve cells before antigen encounter. The increased population size is one reason that antigen challenge in a previously immunized individual induces a larger response than the first immunization in a naive individual.
• Memory cells are able to migrate to peripheral tissues and respond to antigens at these sites. As we discussed in Chapter 3, naive T cells migrate preferentially to secondary lymphoid organs where they respond to antigens for the first time, but memory cells can migrate to virtually any tissue. These differences are related to differences in the expression of adhesion molecules and chemokine receptors. Some memory cells are resident in peripheral tissues and may contribute to recall responses to antigens in these tissues. In addition, memory T cells are less dependent on costimulation than are naive cells, allowing memory cells to respond to antigens presented by a wide range of APCs in peripheral tissues; in contrast, as we have discussed earlier, naive T cells are dependent on antigen presentation by mature DCs in secondary lymphoid organs.
• Memory cells undergo slow proliferation, and this ability to self-renew may contribute to the long-life span of the memory pool. The cycling of these cells may be driven by cytokines. Because of the capacity for self-renewal, memory cells have been likened to stem cells. Although they survive for long periods, memory cells are functionally inactive and have to be restimulated by antigen to become functional effector cells.
• The maintenance of memory cells is dependent on cytokines but does not require antigen recognition. The most important cytokine for the maintenance of memory CD4+ and CD8+ T cells is IL-7, which also plays a key role in early lymphocyte development (see Chapter 8) and in the survival of naive T cells. Predictably, high expression of the IL-7 receptor (CD127) is characteristic of memory T cells. Memory CD8+ T cells also depend on the related cytokine IL-15 for their survival. IL-7 and IL-15 induce the expression of anti-apoptotic proteins and stimulate low-level proliferation, both of which maintain populations of memory T cells for long periods. The ability of memory cells to survive without antigen recognition has been best demonstrated by experiments in mice in which antigen receptors are genetically deleted after mature lymphocytes have developed. In these mice the number of naive lymphocytes drops rapidly, but memory cells are maintained.
The most reliable phenotypic markers for memory T cells appear to be elevated surface expression of the IL-7 receptor and a protein of unknown function called CD27 and the absence of markers of naive and recently activated T cells. In humans, most naive T cells express the 200-kD isoform of the surface molecule CD45 called CD45RA (for “restricted A”), and most memory T cells express a 180-kD isoform of CD45 called CD45RO.
Both CD4+ and CD8+ memory T cells are heterogeneous and can be subdivided into subsets based on their homing properties and functions. However, memory T-cell subsets are best understood for CD8+ T cells and four distinct subsets of CD8+ memory T cells are now recognized.
• Central memory T cells (TCM ) express the chemokine receptor CCR7 and the adhesion molecule L-selectin and home mainly to lymph nodes. They have a limited capacity to perform effector functions when they encounter antigen, but they undergo brisk proliferative responses and generate many effector cells on antigen challenge. Thus, they provide a pool of memory cells that can respond to antigen challenge and develop into effector cells.
• Effector memory T cells (TEM), on the other hand, do not express CCR7 or L-selectin, and they home to peripheral sites, especially mucosal tissues. On stimulation by an anti gen, TEM cells rapidly produce effector cytokines such as IFN-γ or become cytotoxic, but they do not proliferate much. This effector subset, therefore, is poised for a rapid response to exposure to a microbe, but complete eradication of infection may also require large numbers of effectors generated from the pool of central memory T cells. A subset of TEM cells in humans expresses the CD45RA isoform, which is characteristic of naive T cells. This population is called TEMRA cells (T effector memory RA+); whether it has unique functional properties is not known.
• Tissue-resident memory T cells (TRM) are present in various nonlymphoid tissues, do not circulate in the blood, and may provide rapid defense against microbes in the tissues. Most of these cells express high levels of CD69, the molecule that reduces surface expression of S1PR1 (see Chapter 3). As a result, these cells do not respond to the high concentrations of S1P in the lymph and blood, facilitating their retention in tissues.
• Stem-like memory T cells (TSCM) in secondary lymphoid organs express high levels of the TCF-1 transcription factor, and account for the self-renewal and maintenance of central and effector memory T-cell pools, but it is unclear if they are required for the maintenance of the TRM pool.
Memory T cells are also heterogeneous in terms of cytokine profiles. For example, some CD4+ memory T cells may be derived from activated T cells that are not committed to the Th1, Th2, or Th17 phenotype, and when reactivated by exposure to antigen and cytokines, they can differentiate into any of these subsets. Other memory T cells may be derived from differentiated Th1, Th2, or Th17 populations and retain their respective cytokine profiles on reactivation.
