investigated whether the modification of NK lytic activity could be related to differences in the metabolic pattern of activation of NK cells in the elderly. The early signaling events related to hydrolysis of inositol phospholipids were investigated following incubation with K562 target cells and/or CD16 mAb for different times. The data showed a pronounced age-related decrease in the ability to generate total inositol monophosphates and, particularly, inositol trisphosphates by NK cells following K562 stimulation (spontaneous cytolytic activity) together with an attenuated and delayed hydrolysis of phosphatidylinositol bisphosphate, while phosphoinositide turnover was preserved following Fc triggering (antibody-dependent cell mediated cytotoxicity, ADCC). These results confirm that, also in old subjects, different biochemical pathways of activation are involved in NK cells when target or antibody-mediated triggering occurs [99]. Furthermore, other NK studies one of us participated in had demonstrated that decreased cytotoxicity of human NK cells with aging is correlated with a significant decrease in the activity of acid phosphatase in these cells which, in turn, seemed important for maintenance of their cytotoxic activity [100, 101] even if the enzyme does not belong to protein phosphatases which, on the other hand, have an established role in regulating the signal transduction pathways originating with the KIRs [102]. These studies confirmed intracellular molecular changes in NK cells with aging lead to the decrease in cytotoxic activity.
One important aspect of immune aging is the constant antigenic challenge which is mediated by extrinsic and intrinsic stressors [103, 104]. Some extrinsic stressors become intrinsic, as is the case with chronic viral infections which in once infected humans become chronic [105] and whilst maintained under control for most of the lifetime have the tendency to reactivate when immunosurveillance decreases. This is the case for CMV [106]. Most elderly people are infected with CMV, with 80–90% of humans infected at the end of their lifespan. Usually, it is considered that the main defense against this chronic latent infection is the expanded effector memory CD8+ T cells, which may represent a significant percentage of the CD8 population at the end of human life and form part of an immune risk phenotype predicting early mortality in the oldest subjects [43, 106]. Until now, it seems very difficult to disentangle the effects of CMV infection and the aging process which makes CMV infection one of the driving forces of what is called the immunosenescence [106].
Recently, it was shown that NK cells may be influenced by CMV infection [38]. CMV may modulate NK cell functions by means of affecting inhibitory receptor expression whilst avoiding activating receptors. CMV infection induces various NK cell subtypes, but most remarkable is the expansion of mature and dysfunctional (CD56dimCD16+) NK cells which express CD94 and NKG2C [87]. These cells accumulate not only in CMV-infected old individuals but in all CMV infected individuals independently of their age [107]. At the first appreciation, they seem to be protective for the host, especially when they progressively lose the FcRγ and become FcRγ– [108]. These cells express inhibitory receptors for HLA class-I, among them KIR, ILT2 as well as downregulate the activating receptors such as NKp46, NKp30, and acquire CD57 [109]. In infected humans, CMV further induces the memory-like NK cells by the loss of the transcription factor PLFZ, which subsequently will precipitate the loss the SYK, EAT-2, and FcεRγ adapter molecules [110, 111]. Recently, an important role for Tim-3 has been demonstrated in relation to chronic infection and NK cell functions [12]. The loss of Tim-3 is associated with downregulation of IL-12 and IL-18 receptors, returning these NK cells to the quiescent state. These data suggest that CMV induces several progressive changes and adaptation/maturation in NK cells [76]. These adaptive-like NK cells are able to survive and be activated for efficient effector functions on a long-term basis [12]. When they may collaborate with the CMV-specific IgG, they may even more efficiently control CMV infection and in the meantime against other pathogens also [85]. Indeed, the NKG2C+ NK cells expanding as “memory” NK-cells under CMV infection, unlike memory T cells have broad specificity. NKG2C+ NK cells have been observed to expand in response to active hantavirus, chikungunya, HIV, and HBV infections, but only in individuals previously infected with CMV. Recently, a new NK cell subpopulation characterized by the absence of surface expression of CD56 (CD56neg) was described in relation to CMV and EBV coinfections [112]. These cells show decreased cytotoxic activity and IFNγ production, and represent the mature phenotype characterized by low CD57 and KIR expression while lack characteristic features of cell senescence. It is possible that they contribute to the immune dysfunction in aging [112].
Thus, these adaptive-like NK cells can be polyfunctional and affect HSV-1- and influenza-infected cells [44, 113, 114]. They are even more efficient if they express NKp46 or CD2 [113, 115, 116]. These cells are unexpectedly long-lived and are very efficient to combat reactivation by CMV [117]. It was shown that NKG2Chi CD57hi NK cells are also more responsive to NKp46 cross-linking. Both CD16 and NKp46 share the same signaling adaptors, FcRγ and CD3ξ. Fcγ deficiency may enhance the signaling when CD16 and NKp46 have to exclusively use CD3ξ, which contains three immunoreceptor tyrosine-based activation motifs [111]. In the meantime, CMV may also induce NKG2C– NK cells which will be FcRγ– KIR B+ and are also capable of efficient ADCC activity by the activation of the CD3ξ receptor [38, 109, 113, 118, 119], thus sharing the same functional profiles with NKG2C+ NK cells [113]. The downregulation of CD57 in these long-lived mature NK cells is further increasing their reactivity before they lose it [74, 85].