Contrary to the suggestion that susceptibility to new infections in the aged occurs due to insufficient CD8+ T cell immunity as a result of diminished frequencies of naïve CD8+ T cells and/or dysfunctional CD8+ T cell memory 
, we have shown that aged individuals mount CD8+ T cell memory responses to a novel viral agent that are equivalent to young individuals. In fact, extensive analysis of CD8+ T cell functional parameters revealed no relationship between age and the capacity to produce cytokines or mobilize cytotoxic mediators in response to stimulation by peptides derived from viruses responsible for both acute (WNV) and chronic (CMV, EBV) infections despite clear evidence of an immunosenescent phenotype in the bulk CD8+ T cell pool (elevated frequencies of CD28− CD57+ cells and decreased frequencies of CD45RA+ CCR7+ cells relative to the younger members of the cohort). Thus, although the members of our aged cohort displayed expected age-related changes in the composition of the CD8+ T cell compartment, these alterations did not manifest as a defect in functional virus-specific immunity, even when the primary virus infection occurred in old age, as in the case of WNV.
A recent study revealed that infection of middle-aged and old macaques with Rhesus CMV (RhCMV) produced RhCMV-specific CD8+ T cells with comparable functionality in both age groups 
, supporting the concept that anti-viral CD8+ T cell responses may not be dysfunctional in aged individuals. In contrast, immunization with modified vaccinia Ankara (MVA) elicited weaker CD8+ T cell responses in old macaques compared to young macaques 
. While the results of the MVA experiments may seem at odds with our observations, the authors of this latter report employed live vaccinia virus to stimulate MVA-specific CD8+ T cells in vitro
for their functional assays. In contrast, our current report and the report on RhCMV employed synthetic peptides that do not require additional processing for presentation to CD8+ T cells. Since stimulation of CD8+ T cells with live vaccinia virus relies upon the infection, expression and processing of antigen by the cells in the test sample, it is possible that the CD8+ T cell response was intact but antigen presentation by the cells used to present vaccinia antigens in the in vitro
assay were defective in the aged monkeys. The authors argued that DCs were not affected by the age of the monkeys; however, they only investigated a limited number of parameters and they did not examine antigen processing through the classical MHC class I pathway. Therefore, we cannot discount a possible role for defective antigen presentation in their in vitro
stimulation. Another possible explanation for the differences may stem from the nature of the immunogens. MVA is a variant of vaccinia virus that replicates poorly in primate cells, whereas RhCMV and WNV replicate effectively in primate cells. Therefore, effective stimulation of CD8+ T cells responses in the elderly may rely upon the nature of the infectious agent. This will be an important point to consider with regard to vaccine design.
It has been proposed that chronic CMV infection may drive immune senescence due to repeated oligoclonal expansions of CMV-specific CD8+ T cells leading to overpopulation of the memory T cell pool 
and ultimately limiting the ability of the aging individual to combat previously encountered or novel viral infections 
. However, a recent report has suggested that the size of the CD8+ T cell compartment may increase with age to accommodate expanding memory T cell populations without depleting CD8+ T cells with other specificities 
. Interestingly, this phenomenon was not reflected within the peripheral blood where the expanding memory populations increased in frequency at the expense of T cells with other specificities. Rather, the expansion of antigen-specific memory CD8+ T cells was accommodated by increased numbers of CD8+ T cells present within the tissues, suggesting that measures of CD8+ T cell frequencies within the peripheral blood may not accurately reflect the true composition of the CD8+ T cell pool. Although it is relatively easy to measure CD8+ T cells present in the tissues in murine studies, addressing this concept in humans is not trivial. Nevertheless, in light of this recent report, the apparent decline in available naïve CD8+ T cells in the peripheral blood of individuals with evident expansion of CMV-specific CD8+ T cells may not truly reflect a corresponding decrease in the availability of naïve T cells in the lymphoid tissues, where primary responses to viruses are initiated.
Similar to previous reports, we have observed a trend towards higher frequencies of CMV- and EBV-reactive CD8+ T cells in the aged cohort. However, this trend did not achieve statistical significance and not all aged individuals displayed an expanded CMV- or EBV-specific CD8+ T cell pool. Similar results have been reported by others 
. Importantly, in all of these reports, the functionality of the CMV-specific CD8+ T cells did not change with age (the other reports did not investigate EBV-specific CD8+ T cells). It is notable that all of these reports employed functional analyses to define the CMV-specific CD8+ T cells. In contrast, when CMV- and EBV-specific CD8+ T cells were quantified using MHC multimers, it was noted that dysfunctional populations of CMV- and EBV-specific CD8+ T cells accumulate with age based on tetramer staining and IFN-γ production 
. The implications of these dysfunctional cells are unclear as these aged individuals successfully control both CMV and EBV infections and, based on our results, are able to mount effective CD8+ T cell responses to novel infections. We noted a number of mid-aged and aged individuals with frequencies of CMV-reactive CD8+ T cells that represented more than 9% (9 subjects) of the circulating CD8+ T cell pool, but we did not observe any relationship between expanded CMV-specific CD8+ T cells and impaired generation of WNV-specific CD8+ T cells, indicating that CMV expansions do not limit the ability of the host to respond to a novel infection, consistent with the report of Vezys et al. 
Detailed comparison of the functional CD8+ T cell response between the different viruses (WNV, CMV and EBV) revealed interesting differences in functional profiles, corroborating previous reports examining virus-specific CD8+ T cell immunity in humans 
. Striking similarities in both phenotype and cytotoxic profile were observed between memory WNV- and CMV-specific CD8+ T cells, despite the fact that the former is an acute infection and the latter is a chronic infection. The majority of WNV- and CMV-specific CD8+ T cells displayed a phenotype consistent with terminally-differentiated effectors (CD45RA+ CD28−) whereas EBV-specific CD8+ T cells were mostly less differentiated (CD45RA− CD28+) (data not shown). Consistent with the phenotype and the differentiation status, CMV-specific CD8+ T cells produced high levels of GrB and perforin but failed to produce IL-2, whereas EBV-specific CD8+ T cells failed to produce perforin and had less GrB but significantly more IL-2 (); a similar dichotomy in the production of perforin and IL-2 was described in our previous work with a smaller cohort of patients 
. This is consistent with previous reports that show the expression of cytotoxic enzymes is related to cellular maturity, such that CD45RA+/− CD28− cells express high levels of cytotoxicity due to highly differentiated phenotype and CD45RA+/− CD28+ T cells express little cytotoxic attributes 
Collectively, we demonstrate here that aging individuals are capable of mounting polyfunctional memory CD8+ T cell responses to a novel pathogen, which has significant implications for vaccine development for the elderly. Most of our current understanding on the relationship of aging to vaccination has relied upon measurements of antibodies following vaccination and it is clear that the serological response in the elderly is attenuated 
. In striking contrast, our results described herein and in our previous report 
reveal that the aged can mount a robust, polyfunctional CD8+ T cell response to novel pathogens while sustaining a robust polyfunctional responses to chronic infections. Collectively, our data suggest that vaccination in older humans should focus on CD8+ T cell immunity and that live vaccines should be considered as the platform of choice. The results presented here entice our curiosity and desire to better understand the aging immune system for the purpose of developing much needed vaccines for our greatly expanding aging population.