Historically, human NK cells have been identified as CD3-CD56+CD16+/- lymphocytes. More recently it has been established that CD57 expression defines functionally discrete subpopulations of NK cells. On T cells, CD57 expression has been regarded as a marker of terminal differentiation and (perhaps wrongly) of anergy and senescence. Similarly, CD57 expression seems to identify the final stages of peripheral NK cell maturation; its expression increases with age and is associated with chronic infections, particularly human cytomegalovirus. However, CD57+ NK cells are highly cytotoxic and their presence seems to be beneficial in a number of non-communicable diseases. The purpose of this article is to review our current understanding of CD57 expression as a marker of NK cell function and disease prognosis, as well as to outline areas for further research.
Programmed death receptor 1 (PD-1) is an important marker of T-cell exhaustion during HIV-1 infection. Natural killer (NK) cells lose their functional capacity during HIV-1 infection, and PD-1 is expressed on NK cells during other chronic viral and bacterial infections. Here, PD-1 expression was increased on NK cells from both viremic and aviremic HIV-1-seropositive individuals, compared to seronegative controls. However, PD-1 was expressed on a small subset of NK cells and at lower frequency than that observed for CD8+ T cells. PD-1 was also induced on a minor fraction of NK cells and CD8+ T cells after long-term culture with IL-15. Raised levels of PD-1 were associated with limited NK cell proliferation, which may have consequences for their maintenance during chronic HIV-1 infection.
Key Points: HCMV infection in early life is associated with rapid phenotypic and functional differentiation of NK cells. Emergence of CD57+ NK cells is attenuated in children lacking NKG2C.
HIV-1 infection is characterized by increase in inhibitory receptors and loss of activating receptors on natural killer (NK) cells, resulting in loss of cell activity. Exceptionally, for an inhibitory receptor, the proportion of NK cells bearing CD94-NKG2A decreases during HIV-1 infection. It is not understood whether HIV-1 itself or other concomitant infections drive these changes. To investigate the relationship between HIV-1 viraemia and changes in C-type lectin-like receptor expression in NK cells and to investigate the effect of highly active antiretroviral therapy (HAART) on these changes. Three cohorts of patients were studied: (1) before, during and after treatment interruption in aviraemic and viraemic patients receiving HAART (n = 15); (2) HIV-1-positive treatment-naive individuals (n = 13); and (3) HIV-1-positive individuals receiving successful HAART for a minimum of 1 year without interruption (n = 11). Flow cytometry was used to study the expression of NKG2A before and after treatment interruption and to define expanded populations of NK cells in untreated and treated HIV-1-positive individuals. Assays were performed in vitro to assess the cytotoxicity of the expanded populations. Increases in plasma HIV-1 RNA during treatment interruption in aviraemic HAART-treated individuals did not influence the proportion of NK cells carrying the complex CD94-NKG2A. Loss of NKG2A NK cells corresponded to the dramatic expansion of a distinct population of cells expressing a functional activating CD94-NKG2C receptor with skewed expression of killer cell immunoglobulin-like receptor family and natural cytotoxicity receptors. Changes in the NK cell repertoire during HIV-1 infection were not a result of HIV-1 viraemia alone but resembled those associated with concomitant infections.
Human NK cells are activated by cytokines, immune complexes, and signals transduced via activating ligands on other host cells. After vaccination, or during secondary infection, adaptive immune responses can enhance both cytokine-driven and Ab-dependent NK cell responses. However, induction of NK cells for enhanced function after in vitro exposure to innate inflammatory cytokines has also been reported and may synergize with adaptive signals to potentiate NK cell activity during infection or vaccination. To test this hypothesis, we examined the effect of seasonal influenza vaccination on NK cell function and phenotype in 52 previously unvaccinated individuals. Enhanced, IL-2–dependent, NK cell IFN-γ responses to Influenza A/California/7/2009 virus were detected up to 4 wk postvaccination and higher in human CMV (HCMV)-seronegative (HCMV−) individuals than in HCMV-seropositive (HCMV+) individuals. By comparison, robust NK cell degranulation responses were observed both before and after vaccination, due to high titers of naturally occurring anti-influenza Abs in human plasma, and did not differ between HCMV+ and HCMV− subjects. In addition to these IL-2–dependent and Ab-dependent responses, NK cell responses to innate cytokines were also enhanced after influenza vaccination; this was associated with proliferation of CD57− NK cells and was most evident in HCMV+ subjects. Similar enhancement of cytokine responsiveness was observed when NK cells were cocultured in vitro with Influenza A/California/7/2009 virus, and this was at least partially dependent upon IFN-αβR2. In summary, our data indicate that attenuated or live viral vaccines promote cytokine-induced memory-like NK cells and that this process is influenced by HCMV infection.
Studies to develop cell-based therapies for cancer and other diseases have consistently shown that purified human natural killer (NK) cells secrete cytokines and kill target cells after in vitro culture with high concentrations of cytokines. However, these assays poorly reflect the conditions that are likely to prevail in vivo in the early stages of an infection and have been carried out in a wide variety of experimental systems, which has led to contradictions within the literature. We have conducted a detailed kinetic and dose-response analysis of human NK cell responses to low concentrations of IL-12, IL-15, IL-18, IL-21 and IFN-alpha, alone and in combination, and their potential to synergise with IL-2. We find that very low concentrations of both innate and adaptive common gamma chain cytokines synergise with equally low concentrations of IL-18 to drive rapid and potent NK cell CD25 and IFN-gamma expression; IL-18 and IL-2 reciprocally sustain CD25 and IL-18R alpha expression in a positive feedback loop; and IL-18 synergises with FcgRIII (CD16) signalling to augment antibody-dependent cellular cytotoxicity. These data indicate that NK cells can be rapidly activated by very low doses of innate cytokines and that the common gamma chain cytokines have overlapping but distinct functions in combination with IL-18. Importantly, synergy between multiple signalling pathways leading to rapid NK cell activation at very low cytokine concentrations has been overlooked in prior studies focusing on single cytokines or simple combinations. Moreover, although the precise common gamma chain cytokines available during primary and secondary infections may differ, their synergy with both IL-18 and antigen-antibody immune complexes underscores their contribution to NK cell activation during innate and adaptive responses.