SEARCH

SEARCH BY CITATION

Keywords:

  • NK cells;
  • Cord blood;
  • Elderly;
  • CD94/NKG2A;
  • Interferon gamma;
  • cytotoxicity

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Aging is generally associated with an increased predisposition to infectious diseases and cancers, related in part to the development of immune senescence, a process that affects all cell compartments of the immune system. Although many studies have investigated the effects of age on natural killer (NK) cells, their conclusions remain controversial because the diverse health status of study subjects resulted in discordant findings. To clarify this situation, we conducted the first extensive phenotypic and functional analysis of NK cells from healthy subjects, comparing NK cells derived from newborn (cord blood), middle-aged (18–60 years), old (60–80 years), and very old (80–100 years) subjects. We found that NK cells in cord blood displayed specific features associated with immaturity, including poor expression of KIR and LIR-1/ILT-2 and high expression of both NKG2A and IFN-γ. NK cells from older subjects, on the other hand, preserved their major phenotypic and functional characteristics, but with their mature features accentuated. These include a profound decline of the CD56bright subset, a specific increase in LIR-1/ILT-2, and a perfect recovering of NK-cell function following IL2-activation in very old subjects. We conclude that the preservation of NK cell features until very advanced age may contribute to longevity and successful aging.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Aging is a natural postmaturational process associated with increased mortality and morbidity from infectious diseases and cancer, attributable at least in part to defects in immunity (Miller, 1996; Hakim et al., 2004; Pawelec & Larbi, 2008). The complex process of immunosenescence affects both the innate and adaptive arms of the immune system. It has been linked to various factors, including thymic involution, the accumulation of memory T cells, which leads to contraction of the T-cell repertoire, and a decline in B-cell production, reflecting defective humoral immunity (Aw et al., 2007; Hakim & Gress, 2007). At the same time, a variety of evidence indicates that aging exerts significant effects specifically on the innate immune system. Several alterations in the role of natural-killer (NK) cells have been described: both the number of cells and their functions change as people age (Solana et al., 1999; Plackett et al., 2004). Other studies, however, report that aging does not significantly affect NK cytotoxicity (Min et al., 2005; DelaRosa et al., 2006; Solana et al., 2006). The discrepant findings about the effects of age seem due mainly to the different selection criteria for the elderly populations studied; protocols with strict selection criteria that exclude individuals with infections, inflammation or cancer and focus on very healthy older people report the preservation of NK cell cytotoxicity (Mocchegiani & Malavolta, 2004).

NK cells are the first line of defense against infections and developing malignancies. These cells are heterogeneous and differ in their proliferative potential, homing characteristics, functional capacities, and responses to different cytokines. They can be divided into two major subsets based on their relative density of CD56 surface expression. Peripheral blood NK cells are comprised of around 10% CD56brigh and 90% CD56dim NK cell subsets, which are relatively distinct (Caligiuri, 2008); upon activation, CD56bright NK cells proliferate, and produce a wide range of cytokines (e.g. IFN-γ, TNF-β, and IL-10) and chemokines (e.g. MIP-1α and RANTES) but display minimal cytotoxicity activity; in contrast, CD56dim NK cells have little proliferation, produce relatively lower amounts of cytokines, and are highly cytotoxic (Freud & Caligiuri, 2006). Recent experimental evidences have demonstrated that NK development proceeds from a CD56bright to CD56dim phenotype (Chan et al., 2007; Romagnani et al., 2007; Yu et al., 2009). As our understanding of NK cell biology has improved, it has become clear that the balance between inhibitory and activating signals originating from cell-surface receptors dictates their responses. When stimulatory signals outweigh inhibitory ones by a critical threshold, NK cells respond, killing certain infected or transformed cells via perforin/granzyme or death receptor (Fas, TRAIL)-related pathways as well as producing cytokines that influence the host’s immune responses (Vivier et al., 2008). Under normal circumstances of immune self-surveillance, NK cells have inhibitory receptors that recognize MHC class-I molecules as their cognate ligands; these receptors include killer immunoglobulin-like receptors (KIR), LIR-1/ILT-2, and the CD94/NKG2A heterodimeric receptor, as well as LAIR-1, which recognized collagen as ligand (Lebbink et al., 2006). However, cytolysis is not because of the simple absence of MHC class-I molecules but requires an activating receptor. Several of these have been characterized, including NKG2C, NKG2D, the natural cytotoxicity receptors (NKp30, NKp44, and NKp46), and NKp80 (Vivier & Anfossi, 2004; Bottino et al., 2005; Lanier, 2005; Bryceson et al., 2006).

To our knowledge, this study is the first extensive analysis of NK cells from individuals over the human life span from newborns to healthy old people, 25 of them aged from 60 to 80, referred to as old (mean age 68.8 ± 7.2 years) and 30 older than 80 years, referred to as very old (mean age 87.1 ± 4.9 years), passing by 50 adult control subjects (aged 18–60 years), and 20 CB samples. We assessed the phenotypic expression of a large panel of NK receptors in these samples, as well as their functional capacities before and following cytokine stimulation. Our findings provide important clues about NK cell homeostasis and dynamics over time.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Changes in the proportions of lymphocyte subsets across the life span

Flow cytometry measured CD3+ T, CD19+ B, and CD3CD56+ NK subsets in cord blood (CB) and blood from old and very old subjects as well as from adult controls. The frequency of these lymphocyte subsets was similar in CB and peripheral blood from the adult controls, with an exception for the CD3+ T cells, which are significantly decreased (P < 0.001) (Fig. 1). By contrast, for each of these cell subsets, the absolute values were significantly increased in CB samples, when compared to the adult controls, as described (Thornton et al., 2003). In comparison to them, the absolute value of B cells in old subjects was significantly lower (P < 0.05) (Fig. 1a), but the frequency and the absolute values of CD3+ T cells remained together similar to the adult controls (Fig. 1b). Concomitantly, the frequency of CD3CD56+ NK cells was significantly increased but only in the very old subjects (P < 0.05) (Fig. 1c).

image

Figure 1.  Proportion of lymphocyte subsets. Frequency and absolute values of (a) CD19+B cells, (b) CD3+ T cells, and (c) CD3CD56+ NK cells from peripheral blood. (d) FACS profile of CD56bright subpopulation of NK cells gated on CD3CD56+ NK cells. Cells were collected from cord blood (CB, crosses), adult controls (Ctl, squares), old subjects (Old, triangles), and very old subjects (V.Old, diamonds). Horizontal bars represent the median. P values refer to the comparison between the adult controls and either the CB or the old or very old subject groups. *P < 0.05, **P < 0.01, ***P < 0.001.

Download figure to PowerPoint

Human NK cells can be divided into CD56bright and CD56dim populations, based on the cell-surface density of the CD56 molecules; these subgroups present distinct phenotypic and functional properties (Caligiuri, 2008; Poli et al., 2009). The frequency and the absolute value of CD56bright subset are similar in CB samples and adult controls but decrease progressively with age (P < 0.05 for old and P < 0.001 for very old subjects, in frequency) (Fig. 1d).

NK cell expression of inhibitory and activating receptors across the life span

We further compared the phenotypic properties of a more restricted NK cell subset by gating out the CD56bright NK cells (Supporting Information Fig. S1). Instead, we focused our study on CD56dim cytotoxic NK cells, analyzing the expression of a large panel of activating and inhibitory receptors on them (Fig. 2 and Supporting Information Figs S2 and S3). Expression of the inhibitory receptor CD94/NKG2A was sharply higher in CB samples (median 80.6 ± 7.7%) than in samples from adult controls (median 45.9 ± 15.5%) (P < 0.001), which in turn was similar to that in both categories of older subjects (Fig. 2a). In contrast, the proportion of NK cells expressing the activating NKG2C receptor, which recognizes the same HLA-E ligand, as the NKG2A receptor (Braud et al., 2008), was low and similar in all samples. By contrast, expression of NKG2D was significantly lower in CB (P < 0.01) than in samples from adult controls and elderly (Fig. 2a). The expression of NKp30 and NKp46, two other major specific activating NK receptors, was significantly higher in CB samples (P < 0.05 and P < 0.01, respectively) than in any of the adult samples, whether adult controls, old, or very old subjects (Fig. 2b).

image

Figure 2.  Pattern of receptor expression on CD3CD56dim NK cells. Peripheral blood was collected from cord blood (CB, crosses), adult controls (Ctl, squares), old subjects (Old, triangles), and very old subjects (V.Old, diamonds). (a) Expression of c-lectin NKG2A, NKG2C, and NKG2D receptors (b) Expression of activating NKp30, NKp44, and NKp46 receptors, (c) Expression of killer cell Ig-like receptors (KIR) including KIR2DL1/DS1, KIR2DL2-3/DS2, and KIR3DL1/DS1, and (d) Expression of other NK cell markers, including ILT-2, CD57, and LAIR-1. All cell-surface receptors were gated on CD3CD56dim NK cells. Horizontal bars represent the median. P values refer to the comparison between the adult controls and the CB or the old or very old subject groups. *P < 0.05, **P < 0.01, ***P < 0.001.

Download figure to PowerPoint

We next looked at the expression of KIRs, including KIR2DL1/DS1 and KIR2DL2/DL3/DS2, which bind group-2 and group-1 HLA-Cw ligands, respectively, and KIR3DL1/DS1, which recognizes HLA-Bw4 (Moretta & Moretta, 2004). Expression of these KIR markers in adult controls was higher than in CB but similar or slightly lower than in samples from old and very old subjects (Fig. 2c). More importantly, as previously described (Sundström et al., 2007; Béziat et al., 2009), LIR-1/ILT-2, an inhibitory receptor that recognizes a broad range of classical MHC class I molecules on surrounding cells, was expressed at much lower levels in newborns than in adult blood donors. Indeed, less than 8% of NK cells from CB expressed LIR-1/ILT-2, compared with 37.7 ± 17.1% in the adult controls (Fig. 2d). Moreover, the proportion of NK cells expressing LIR-1/ILT-2 increased significantly with age: 50.8 ± 18.9% of NK cells from old subjects and 60.2 ± 11.6% of those from the very old were positive for LIR-1/ILT-2 (Fig. 2d). Similarly, expression of CD57, a useful marker for terminal differentiation on NK cells, was significantly increased on NK cells with age (Fig. 2d), as previously described (Tilden et al., 1986). Finally, these samples were indistinguishable according to age for cell-surface expression of the other NK receptors tested, including NKp44, NKp80, 2B4, and LAIR-1 (Fig. 2 and data not shown).

The rate of activated NK cells in these different age groups was assessed by flow cytometric determination of the early activation marker CD69 (Fig. 3). As Fig. 3a shows, in CD3CD56dim NK cells, the level of this activation marker was similar in CB samples and adult controls but was significantly higher in very old subjects (P < 0.01). Results were similar in CD3CD56bright immunoregulatory NK cells (Fig. 3b), suggesting a close relation between the activation status of NK cells and their course over the human life span.

image

Figure 3.  Cell-surface expression of activation NK cell marker. Peripheral blood was collected from cord blood (CB, crosses), adult controls (Ctl, squares), old subjects (Old, triangles), and very old subjects (V.Old, diamonds). Expression of the early CD69 activation marker was tested on the (a) CD3CD56dim and (b) CD3CD56bright NK cells. Horizontal bars represent the median. P values refer to the comparison between the adult controls and the CB or the old or very old subject groups. *P < 0.05, **P < 0.01, ***P < 0.001. (c) Representative expression of CD69 on CD3CD56+ NK cells from cord blood (CB), adult controls (Ctl), old subjects (Old), and very old subjects (V.Old). Percentage of cells is shown in each quadrant.

Download figure to PowerPoint

Functional analysis of NK cells across the life span

We next assessed the overall functional ability of NK cells at different ages. One of the main functions of NK cells is the capacity to synthesize and release some cytokines, especially IFN-γ, which participate in the initiation of the Th1-dependent adaptive immune response. The level of IFN-γ after treatment with IL-12 and IL-18 was very significantly higher in CB (P < 0.001) than among the adults, for whom the proportion of IFN-γ+ cells was similar regardless of age (Fig. 4a,b). Of note, the increased proportion of IFN-γ production in CB is independent of the CD56dim and CD56bright subsets, both producing IFN-γ following IL-12 plus IL-18 stimulation (Fig. 4a).

image

Figure 4.  IFN-γ production and cytolytic activities of NK cells. Peripheral blood was collected from cord blood (CB), adult controls (Ctl), old subjects (Old), and very old subjects (V.Old). (a) Intracellular expression of IFN-γ was tested on CD3CD56+ NK cells from PBMC after incubation with IL-12 and IL-18. Percentage of cells is shown in each quadrant. (b) Cumulative analysis of IFN-γ production into 10 independent samples from each group of subjects. A black bar inside the box-and-whiskers plots indicates the median. (c) NK cytolytic activity of IL-2 activated NK cells was tested against K562 target cells by a standard 51Cr assay. A black bar inside the box-and-whiskers plots indicates the median. (d) cytotoxicity (left panel) and degranulation response (right panel), determined by a standard 51Cr assay, and CD107a expression, respectively, were tested with nonactivated NK cells against K562 target cells. Results are shown at an E/T ratio of 40/1 and 1/1, for the direct lysis and the degranulation, respectively. A black bar inside the box-and-whiskers plots indicates the median. (e) Representative expression of CD107a on nonactivated CD3CD56+ NK cells from cord blood (CB), adult controls (Ctl), old subjects (Old), and very old subjects (V.Old) without (upper panel) or with K562 target cells (lower panel). Percentage of cells is shown in each quadrant. (f) Expression of CD16 on CD3CD56dim NK cells. Horizontal bars represent the median. (g) Degranulation of nonactivated NK cells was tested against Raji cells in the absence (open bars) or in presence of 1 μg/ml of anti-CD20 mAb (hatched bars). Degranulation was shown at an E/T ratio of 1/1. A black bar inside the box-and-whiskers plots indicates the median. P values refer to the comparison between the adult controls and the CB or the old or very old subject groups. *P < 0.05, **P < 0.01, ***P < 0.001.

Download figure to PowerPoint

Another NK-cell function is the capacity for direct cytotoxicity against HLA-class I-negative K562 sensitive-target cells. The cytotoxic potential of these cells was compared after IL-2 activation; NK cells from CB killed at a rate similar to that of samples from adult controls. This cytolytic activity was also well preserved in the older populations (Fig. 4c). Surprisingly, response was highly variable in the younger of the old groups, with some samples showing poor NK cytolytic capacity (Fig. 4c). This suggests that NK cells from each group of patients have not the same responsiveness to IL-2 activation, but also that well-preserved NK-cell activity later in life might be interpreted as a factor of longevity. Of note, in absence of IL-2 activation, a significant decrease in NK lysis was observed in CB (P < 0.05), when compared to the controls. In addition, the older subjects almost achieving statistical significance, in accordance with previous studies (Ogata et al., 1997; Ravaglia et al., 2000; Bruunsgaard et al., 2001) (Fig. 4d). These data were confirmed by their lower capacity to spontaneously degranulate in the presence of K562 target cells (Fig. 4d,e). Together, these results strongly suggested that older subjects have lower cytolytic capacity but importantly, we observed a perfect recovering of NK-cell function following IL-2 activation in these very old subjects.

Finally, another important NK cell function is the mediation of antibody-dependent cellular cytotoxicity (ADCC). Surface expression of FcγRIIIA/CD16 on NK cells did not decrease significantly with age, from birth to the oldest subjects (Fig. 4f). Nonetheless, as Fig. 4g shows, degranulation capacity of NK cells against anti-CD20-pulsed Raji target cells was significantly poorer in CB samples (P < 0.01) than in adult controls. The proportion of CD107a+ NK cells from CB was 3.3-fold lower than in adult controls, in the presence of anti-CD20 mAb (Fig. 4g). In contrast, this functional capacity of NK cells appeared intact in the older individuals, which suggests that once adulthood is reached, age does not affect this NK-cell function, as described (Mariani et al., 1998).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

In this study, we report phenotypic and functional changes of NK cells with age from birth through very old age. In line with previous data, we observed that NK cell frequency and absolute value tend to increase with age. Interestingly, Remarque & Pawelec (1998) report that among older people, those with a low NK cell count have a risk of mortality three times higher than those with high NK cell counts, in the first 2 years of follow-up. Accordingly, the preservation or increase in the NK cell population with age may be interpreted as a factor of longevity.

Our findings also show that the frequency and absolute value of CD56bright NK cells gradually declines with age. Concomitantly, the CD56dim NK cell subset occupies an increasingly greater portion of the lymphocyte pool in old subjects. Together, these results suggest that the increase in NK cell frequency with age results from expansion of the mature CD56dim rather than of the immature CD56bright NK subset. These results were consistent with previous reports showing a profound decrease in CD56bright NK cells in elderly subjects (Borrego et al., 1999; Chidrawar et al., 2006). This phenomenon may be a compensatory mechanism to offset a possible reduction in per-cell cytotoxicity. However, we cannot rule out the possibility that a decline in CD56bright NK cells contributes to the development of immune senescence, as this population is critical to the cytokine response of the innate immune system (Caligiuri, 2008). In addition, CD56bright NK cells play an important role in the activation of dendritic cells (Vitale et al., 2004; Moretta et al., 2006) and also interact reciprocally with monocytes, thereby promoting inflammation (Dalbeth et al., 2004).

Several age-related changes in the expression of NK cell receptors reinforced this notion of progressive maturation of NK cells, from CB to very old people. We showed that NK cells from CB display specific features that differ from adult NK cells and that some of these features, including low expression of KIR and LIR-1/ILT-2 and high expression of CD94/NKG2A and natural cytotoxicity receptors, are associated with a less mature stage of differentiation. However, we observed, as have others (Rocha et al., 2000; Béziat et al., 2009), that these NK cells from CB could rapidly become fully mature after hematopoietic transplantation. Interestingly, we showed a reciprocal age-related change in NK receptors for MHC class-I, between CB and adults NK cell samples, including a decrease in CD94/NKG2A expression and a reciprocal increased in KIR. However, contrasting with Lutz et al. (2005), we do not detected this shifting in mature NK cells between adult controls and elderly subjects; this discordance could be du to the age-diversity of the elderly subject tested in both studies.

Interestingly, the features of the inhibitory receptor LIR-1/ILT-2 are highly specific at different stages of life. This receptor has a broad recognition pattern for classical MHC class-I (HLA-A, HLA-B, HLA-C) and nonclassical MHC class-I (HLA-G) antigens. It has been previously reported that approximately 30% of mature NK cells express this receptor (Young et al., 2001; Sundström et al., 2007). Although the proportion of LIR-1/ILT-2 in CB is very low, the proportion of NK cells expressing this marker is similar in 5-year-olds and adults (Sundström et al., 2007; Béziat et al., 2009). More surprisingly, the proportion of LIR-1/ILT-2+ NK cells continues to increase, albeit gradually, with age in adults and reaches a level twice as high in older subjects. LIR-1/ILT-2 also recognizes UL18, a CMV protein as ligand (Kim et al., 2004), suggesting that the level of this receptor could be associated to the CMV status of the subjects. However, analysis of the serum CMV IgM and IgG antibody titer levels revealed an absence of CMV-reactivation in all subjects, and then no relationship with the frequency of LIR-1/ILT-2+ NK cells (data not shown). Alternatively, Wagner et al. (2007) showed that exposure to locally produced cytokines, such as IL-15, can induce NK cells to express LIR-1/ILT-2. Alteration in the cytokine milieu with age might thus explain the increased LIR-1/ILT-2 expression observed here.

On the other hand, expression of the CD69 activation marker increases significantly with age and reached 75% in a few very old subjects. In contrast and consistently with earlier studies (Sundström et al., 2007), less than 10–15% of NK cells from CB and from adult subjects expressed this marker. Expression of CD69 by NK cells may certainly reflect an in vivo activation state, but we cannot rule out the possibility that expression of this marker also plays a role in the retention of cells in secondary lymphoid tissues (Shiow et al., 2006). Together, these data clearly suggest that the shape of the NK-cell receptor repertoire changes over the course of our lives, reflecting the completion of maturation and in relation to environmental factors.

The predominance of CD56dim in the mature NK cell subset leads to both a phenotypic and a functional shift in the maturity status of NK cells during aging. The issue is substantially more complex, however, because NK cell activity is controlled by cytokines and chemokines, which also change with aging. In these conditions, NK cells from CB showed some specific functional features, in particular a profound increase in IFN-γ production and a decreased capacity for effective degranulation in presence of anti-CD20-pulsed Raji target cells, related to their immature phenotype. These findings corroborated earlier reports (Sundström et al., 2007; Béziat et al., 2009). Accordingly, although several studies have investigated the effects of age on NK cell function, considerable controversy remains about whether advanced age affects NK-cell functions adversely. They are several data supporting that changes in NK cells can be associated with longevity but also with health. For this reason, our study applied restrictive health selection criteria including no infectious, malignant, or autoimmune diseases during the 6 months before the study, and without acute illnesses at the time of the sampling; however, the paucity of all available antecedent parameters thus represents a limit of our study.

The ability to produce IFN-γ was modestly impaired in NK cells from older subjects, compared with adult controls. Similarly, in absence of IL-2 activation, we have observed a decrease in NK lysis in those older subjects. In line with these data, age-associated alterations in NK cell functions have been reported in disease states, suggesting a higher incidence of infections, higher rate of atherosclerosis onset, and increased nutritional deficiencies in those older subjects (Ogata et al., 1997; Ravaglia et al., 2000; Bruunsgaard et al., 2001). Importantly, we observed a perfect recovering of NK-cell function from the very old subjects, following IL2-activation. Interestingly, in this condition, a relative decrease in NK cell function was more evident in some of the old subjects, aged from 60 to 80 than in the very old subjects, that is, older than 80. This is consistent with previous studies (Mariani et al., 1999; Mocchegiani et al., 2003). Together, these results suggest that the preservation of effective NK-cell function in early old age (before 80 years) may make it possible to avoid some age-related diseases and thus reach a healthy old age.

In summary, the evidence from this study shows phenotypic and functional upheavals in NK cells during immunosenescence, including a preferential accumulation of mature NK cells. We can speculate that these changes may be caused by environmental factors, including telomeric loss, oxidative damage, and DNA defects, which may influence each other and create a sort of vicious spiral associated with the passage of time.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Twenty CB samples were provided by the obstetrics department of Pitié-Salpêtrière Hospital (Paris, France), and 50 blood samples from adult blood donors, referred to as adult controls (younger than 60 years), from the Etablissement Français du Sang (EFS). The rheumatology and gerontology departments of Pitié-Salpêtrière or Charles Foix Hospitals furnished 55 samples from patients aged 60 or older: 25 aged 60–80 and referred to here as old (mean age 68.8 ± 7.2 years) and 30 older than 80 years and referred to as very old (mean age 87.1 ± 4.9 years). All volunteers affirmatively stated, and their medical records confirmed, that they had no infectious, malignant, or autoimmune diseases during the 6 months before the study, and without acute illnesses at the time of the sampling. Informed consent was provided in all cases in compliance with the Ethics Committee guidelines before peripheral blood samples were collected for the study.

Multi-color FACS analysis was performed on freshly harvested blood cells, as described (Béziat et al., 2009). The flow cytometric gating strategy was shown in Supporting Information Fig. S1. NK cells were analyzed on the lymphocytes gate, after staining with an appropriate antibody cocktail: anti-CD3 (UCHT1), anti-CD56 (N901), anti-CD16 (3G8), anti-KIR2DL1/KIR2DS1 (EB6B), anti-KIR2DL2/KIR2DL3/KIR2DS2 (GL183), anti-KIR3DL1/KIR3DS1 (Z27), anti-CD159a/NKG2A (Z199), anti-NKG2D (ON72), anti-CD336/NKp44 (Z231), anti-CD335/NKp46 (BAB281), anti-CD85j/ILT2 (HP-F1), anti-2B4 (C1.7), and anti-CD69 (TP1.55.3) from Beckman Coulter; anti-NKG2C (134591) and NKp80/KLFR1 (239127) from R&D Systems; anti-CD337/NKp30 (AF29-4D12) from Miltenyi Biotech; and anti-CD57 (NK1) and anti-CD226/DNAM-1 from Becton Dickinson. Each antibody was tested for its efficiency in presence of specific negative and positive controls. FACS lysing solution (Becton Dickinson, le pont de Claix, France) was used to lyse erythrocytes. At least 50 000 lymphocytes were analyzed on a FC500 (Beckman Coulter, Villepinte, France) and/or FACSCanto I (Becton Dickinson).

To stimulate IFN-γ production, PBMC were incubated overnight in the presence of IL-12 (10 ng mL−1) and IL-18 (100 ng mL−1) (R&D Systems, Lille, France), as previously described (Nguyen et al., 2008). Cells were fixed and permeabilized with cytofix/cytoperm kit (Becton Dickinson) and then stained with anti-IFN-γ mAb (B27; Becton Dickinson), as described (Nguyen et al., 2005).

Direct cytolytic activity of whole NK cells was assessed on PBMC, in a standard 4-h 51Cr-release assay or in a degranulation assay by the detection of CD107a, cultured in the absence or in the presence of 1000 IU mL−1 of Proleukin (Chiron, Suresnes, France) for 48 h, against K562 target cells (Alter et al., 2004). Degranulation of NK cells in presence of anti-CD20-pulsed Raji cells was performed according to methods previously described (Béziat et al., 2009). Briefly, nonactivated PBMC were resuspended with Raji target-cells at an effector:target cell ratio of 1:1 in the presence of anti-CD107a mAb (H4A3; Becton Dickinson), plus 1 μg mL−1 of anti-CD20 mAb (Rituximab; Roche, Strasbourg, France). After 1 h of incubation, monensin (Sigma Aldrich, Lyon, France) was added at 6 μg mL−1 for an additional 3 h of incubation (Béziat et al., 2009).

All statistical analyses were performed with Prism 5 software (GraphPad Software, San Diego, CA, USA). Intergroup comparisons were assessed with the nonparametric Krustal–Walis test, with the Dunns postanalysis test to define the significance between results from independent group of subjects. Significance defined by P less than 0.05 with a two-tailed test. *P < 0.05, **P < 0.01, ***P < 0.001.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

The research was supported in part by the French Ligue contre le Cancer (RS08/75-4). We thank Cecile Poulain (Service de Rhumatologie, Hôpital Pitié-Salpêtrière, Paris, France) for the recruitment of aging subjects, and Claire Deback (Service de Virologie, Hôpital Pitié-Salpêtrière, Paris, France) for analysis of CMV status in subjects. We also thank the personnel from the Etablissement Français du Sang (EFS) for the healthy adult blood samples and those from the Department of Gynécologie-Obstétrique at the Pitié-Salpêtrière hospital (Paris, France) for the cord blood samples.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Fig. S1 Flow cytometric gating strategy.

Fig. S2 Representative pattern of receptor expression on CD3CD56+ NK cells from cord blood (CB), adult controls (Ctl), old subjects (Old), and very old subjects (V.Old).

Fig. S3 Representative pattern of receptor expression on CD3CD56+ NK cells from cord blood (CB), adult controls (Ctl), old subjects (Old), and very old subjects (V.Old).

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
ACEL_584_sm_figS1.tif778KSupporting info item
ACEL_584_sm_figS2.tif1460KSupporting info item
ACEL_584_sm_figS3.tif1497KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.