Gerontology 1992:38:41-49
Jolanta Mysliwskaa Andrzej Mysliwskia P/o/r Romanowskia /aci'A' Bigdaa Danuta Sosnowskaa 7(?rry Foersterb
Monocytes Are Responsible for Depressed Natural Killer (NK) Activity in both Young and Elderly Low NK Responders
3 Department of Histology, Medical School. Gdansk, and b Geriatric Center. Gdynia-Witomino. Poland
Key Words Human NK activity Age-related differences Monocytes
Abstract Two age groups - young (19-35) and elderly (70-91) - were compared with respect to natural killer (NK) cytotoxic activi ty. In both groups, low and high NK responders could be dis tinguished. Low NK responders constituted about 70% of all elderly and 40% of young individuals. The differences in the magnitude of NK activity among the young and elderly groups could only be observed when peripheral-blood mononuclear cells but not peripheral-blood lymphocytes were used as effec tor cells in a 51Cr release assay. Experiments with removal or addition of graded numbers of monocytes showed that these cells were responsible for the low level of NK activity in both the young and elderly low NK responders.
The ageing process is claimed to be associ ated with a progressive decline of T- and Bcell-dependent immune responses [16]. Stud ies of age-associated alterations in the expres sion of natural killer (NK)-cell-mediated cytolysis revealed all possible changes, i.e. no sig nificant change in the intensity of NK activity [2,15, 18], or increase [6,9, 10] or decrease [7, 13, 19] in the magnitude of response of NK cells.
The age-related increase in NK activity [9] as well as the expansion of the NK cell com partment found in old. healthy individuals [10, 12] may lead to the conclusion that the well-preserved NK cell system in the elderly plays a compensatory role for a decreased function of the T and B cell compartments. The NK cell compartment may then be im portant for survival.
This work was supported in part by the II Polish-AmericanFundM.Sklodowska-Curic.
Dr. J. Myskwska Department of Histology Medical School 7 0 -2 10 Gdansk (Poland)
Received: June 16.1991 Accepted: August 23. 1991
© 1992 S. Kargcr AG, Basel 0304-324X/92/ 0382-0041$2.75/0
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Introduction
Subjects and Methods A group of 35 old volunteers (age range 70-91: 6 men, 29 women), inhabitants of the Old People’s Home in Gdynia-Witomino, was examined. They were under constant medical supervision and regularly un derwent twice a year a general medical examination. An analysis of the following blood parameters was rou tinely done: erythrocyte sedimentation rate: hemoglo bin level; mean corpuscular volume: leukocyte count with differentiation: levels of glucose, urea, lipids, aspartate aminotransferase and alanine aminotransfer ase, and protein electrophoresis. Urinalysis comprised protein, glucose, ketones and sediment. All people admitted to the study were not suffering from chronic infections, malignant disease, lymphoproliférative dis orders and dementia. Only those volunteers were qual ified who had their last acute infection at least 2 months before examination and had not been hospital ized for the last 6 months. Volunteers taking regularly medications, except for vitamins, were not admitted. The control group consisted of 30 young adults medical students and staff members (age range 19-35; 5 men. 25 women) - who were qualified after a careful medical examination. The same blood and urine pa rameters, as in the case of the old group were deter mined. Young volunteers, like elderly, were not suffer ing from acute and chronic infections, malignant dis ease and lymphoproliférative disorders.
42
Preparation o f PBMC Human PBMC were obtained by centrifugation of heparinized blood samples on Lymphoprep gradient (Nyegaard, Norway). After 3 washings with phosphatebuffered saline (PBS), the cells were suspended in RPMI 1640 medium (Gibco. UK) supplemented with 5% fetal calf serum (FCS: Flow Labs.. USA). A me dium composed in such a way will be referred to as ‘medium’. In order to obtain PBL. PBMC samples were incu bated on plastic Petri dishes for 1 h in a humidified atmosphere containing 5% CCH at 37 °C. The plasticnonadherent lymphocytes were separated from plasticadherent monocytes. Different concentrations of monocytes in mononu clear cell suspension were obtained by dilution of PBMC with PBL suspension. Monocytes were identi fied on the basis of the pattern of their acid a-naphthvlesterase (ANAE) activity. B. T and NK cells were depleted from PBL by C'mediated lysis. PBL were incubated for 30 min on ice with one of the following monoclonal antibodies (MoAB): BMA 0130 (anti-CD 19): BMA 081 (antiCD8): BMA 041 (anti-CD4). or BMA 070 (anti-CD 16: Bchringwerke, FRG). After incubation cells were cen trifuged, washed with PBS and incubated with an appropriate amount of guinea pig serum at 37 °C for 30 min. Depletion of a given population was checked with an immunoperoxidase method. In control experi ments, guinea pig serum alone or the irrelevant MoAb BMA 0220 (anti-Reed-Stcrnberg cells: Bchringwerke) was incubated with effector cells which were processed identically as those treated with the relevant MoAb. Preparation o f Target K562 Cells The K.562 human erythroleukemia cell line was maintained in RPMI 1640 medium (Gibco) contain ing 10% FCS (Flow Labs.). 100 pg/ml streptomycin, 100 lU/ml penicillin and 2 m.V//.-glutamine (Gibco). Assessment oJ'NK Cell Activity Single-Cell Cytotoxicity Assay. Two million of both effector (PBMC. PBL) and target (K562) cells sus pended in medium were mixed, centrifuged at 400 g, incubated for 15 min at 37 °C and resuspended. In some experiments, CD16+ cells were marked in the suspension of effector cells before mixing them with K562 cells. PBMC were incubated for 30 min on ice with an appropriate amount of BMA/070 MoAb (anti-CD 16), centrifuged and washed 3 times in PBS. After resuspension cells were incubated for 15 min with an appropriate amount of magnetic Dvnabeads M450 coated with goat antimouse IgG (Dvnal. Nor
MySliwska/MySliwski/Romanowski/ Bigda/Sosnowska/Focrstcr
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
For the determination of NK cytotoxic ac tivity, a 51Cr release assay with K562 cells as targets is routinely applied. As effector cells, some authors use peripheral-blood lympho cytes (PBL) [6, 13] whereas others utilize pe ripheral-blood mononuclear cells (PBMC) [2, 18], The fact that under different experimen tal conditions monocytes can exert various effects on the cytotoxicity of NK cells [5, 8] implies that their presence in the cytotoxic test could be of relevance for the differences in NK activity between young and old per sons. In the present study, we compare NK activity in young (19-35 years old) and el derly (70-91 years old) people. Special atten tion is paid to the differences in NK activity due to the utilization of PBMC and PBL as effector cells.
Staining o f CD3*, CD4*. CD8* and CD 16* cells by Immunoperoxidase Method Conjugates were formed between effector and tar get cells and transferred onto the glass slides. Slides were fixed with acetone. To detect a given cell popula tion, slides were coated with MoAb BMA 030 (antiCD3; diluted 1:5 in PBS/BSA), BMA 040 (anti-CD4), MBA 081 (anti-CD8) or BMA 070 (anti-CD 16). After 30 min of incubation, slides were rinsed with PBS and finally coated for another 30 min with peroxidase-con jugated rabbit antimouse Ig serum (Dako, Denmark, diluted 1:100). After rinsing with PBS. the peroxidase reaction was carried out. The incubation mixture con sisted of 6 mg diaminobenzidine (Sigma, USA) dis solved in 10 ml, of PBS and 5 pi of 30% HiCK The reaction was stopped with distilled water. Slides were counterstained with hematoxylin and finally im mersed with glycerogei. Activity of endogenous peroxidase was abolished by incubation of slides with 1% H2O 2 in methanol for 15 min. Statistical analysis was performed by means of a paired t test.
Results NK Activity o f PBMC o f Elderly Persons
In each experiment. NK activity was deter mined in pairs of young and old donors and the arithmetic mean of the values of the young persons was assumed as index = 1.0. The val ues of 51Cr release of PBMC of the elderly were calculated in relation to the values of the young subjects. As shown in figure 1, the index expressing NK activity of elderly people was 0.543 ± 0.205, which indicates that the intensity of their NK activity was about half that charac teristic of young people. The difference was statistically significant (p < 0.001). Cellular Composition o f Conjugates o f Old and Young Persons
Besides cytolytically active NK effectors characterized by the presence of CD 16 anti gen [11, 14], representatives of the T cell lineage not exhibiting cytotoxic activity be long also to the cells forming conjugates with NK-sensitive targets [12], It is thus possible that conjugates of the elderly were enriched in the nonlytic lymphocytes. To check whether this was really the case, an analysis of the cells in the conjugates was performed. It was found that about 98% of conjugates consisted of one target and one effector cell. The percentage of cells forming conjugates was similar in both age groups: 15.4 ± 1.3% in the young and 14.8 ± 3.2% in the elderly. There were no sta tistically significant differences between the participation of conjugating cells carrying CD 16. CD3, CD4 and CD8 antigens (fig. 2). As follows from our results. PBMC of old individuals formed as many conjugates with K562 targets as those of young subjects. Moreover, conjugates of old and young indi viduals were not distinctly composed with respect to the cells expected to be both lytic and nonlytic.
43
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
way). The suspension of effector and target cells was gently mixed with an equal volume of warm (3942 °C). liquified agarose (Windsor Labs., UK) to give a final concentration of agarose of 0.5%. The mixture was then layered on plastic Petri dishes (Technomcd, Poland). After solidification of agarose, medium was poured on the agarose surface. The cells were incu bated for 3 h at 37 °C in a humidified atmosphere con taining 5% COi. After incubation dishes were stained with 1% solution of trypan blue. The percentage of tar get-binding cells with dead targets was counted. The cytotoxic activity of cells in conjugates (cytotoxic in dex; Cl) was calculated according to the formula of Targan and Dorey [ 17): Cl = percentage of dead targets in conjugates - (fraction of spontaneous dead targets X percentage of dead targets in conjugates). 5lCr Release Assay. The samples, run in triplicate, contained a mixture of 2 X 104 K562 cellsand 5 X I05 effector cells. The volume of samples was adjusted to 0.4 ml. The samples were incubated in 5% COi at 37°C for 4h. then centrifuged at 2,000 rpm for 10 min. The 0.25-ml supernatants were removed and counted for gamma radioactivity in an LKB gamma counter. The specific cytotoxicity was computed by the following formula: percent chromium release = (exper imental release - spontaneous release)/(maximal re lease - spontaneous release).
1 . 2-1
Young
Elderly
Fig. 1. NK activity of PBMC of young (n = 30) and elderly (n = 35) individuals measured in a 51Cr release assay. In each experiment NK activity was determined in pairs of young and old donors, and the arithmetic mean of the values of young persons was assumed as index = 1.0. Values of the elderly were calculated in relation to the values of the young subjects. The differ ence between the values of the young and elderly was statistically significant (p < 0.001).
Fig. 2. Percentage of conjugates formed by CD 16*. CD3*. CD4‘ and CD8* cells in young (ED and elderly ( B individuals. The differences between the values of the young and elderly in each conjugate group were not statistically significant (p > 0.05).
Cytotoxic Activity o f CD I6+Cells Expressed at a Single-Cell Level
44
Fig. 3. Cytotoxic activity as assessed in a single-cell cytotoxicity assay with PBMC, and cells bearing (CD16*) and lacking (CDI6-) CD16 antigen. The dif ferences between the activity of CDI6+ and CD 16populations were statistically highly significant (p < 0.0001). Young subjects: 0 « elderly subjects.
values for the two age groups, pointing that cells not carrying the CD 16 marker were also not responsible for the differences in NK activity between young and old participants either. Since the anti-CD 16 MoAb used as a marker of NK cells in a single-cell cytotoxic ity test could have an impact on NK activity [1], in the next experiment a single-cell cyto-
MySliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerster
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Thus, although CD16+ cells of the elderly were contained in conjugates in a similar pro portion to those of the young subjects, the cytotoxic activity of the elderly was reduced. This fact implied that some CD 16+cells of the elderly might have been devoid of cytolytic activity. This possibility was tested in the next experiment in which CD 16-antigen-bearing cells, regarded as the strongest cytolytically active NK population [11], were marked by an immunomagnetic method and tested in a single-cell cytotoxicity test in agarose. As shown in figure 3. almost all NK activ ity of both age groups resided in the subpopu lation of CD16+ cells, confirming that these cells were the actually cytolytically active ones. CD 16+cells of the elderly people were as active as those of the young group (p > 0.05). The Cl of the CD 16" population, though markedly lower, was also in a similar range of
2.0-,
-C D 22
¡MoAb -CD4 -C D 8 -C D 16 PBMC
PBL
Fig. 5. Effect of depletion of different cell populations from PBMC on NK activity of elderly people as measured in a 5lCr release assay. NK activity is shown after depiction of CD22+ cells (-CD22), in PBMC incu bated with C', without MoAb (C') in PBMC incubated with an irrelevant MoAB (iMoAb) after depletion of CD4' cells (-CD4). CD8* cells (-CD8) and CD16- cells (-CD16). in nontreated PBMC and in PBL (PBL). The index of increase was calculated as follows: (NK activity after dcpletion/NK activity before depletion) - l.O. Depletion of monocytes pro duced a statistically significant increase (p < 0.001) in NK activity.
toxicity test was carried out on the nonmarked PBMC and again the Cl of both groups was compared. Cl of the young was 21 ± 5.6 and that of the elderly 19.5 ± 7.1. This indicated that anti-CD 16 MoAb did not af fect the outcome of a single-cell cytotoxicity test and reinforced the former finding that the cytolytic effectiveness of all cytolytically ac tive cells was equal in both age groups. Taken together, these results demonstrate that the defect of NK activity observed in the group of elderly people, measurable in a 5lCr release assay, does not depend on the reduced number of active NK cells. Percentage o f CD 16* Cells in Blood As a reduced number of cytotoxically ac tive NK cells in conjugates did not account for the lower cytotoxic activity of PBMC of
the elderly, it was possible that elderly people had less CD16+ cells in blood and therefore a reduced general cytotoxic potential of their PBMC, as reflected in a 5lCr release assay. This possibility was tested in the next experi ment. It appeared, however, that both age groups possessed a comparable percentage of CD 16+ cells in the blood (fig. 4). Effect o f Depletion o f Different Celt Populations on NK Activity
The discrepancy between the tendency of the NK activity values of old people to drop when measured in the 5lCr release assay and to be on the same level as those of the young when assessed in a single-cell cytotoxicity as say suggests that under the conditions of the 51Cr release assay a certain population of cells present in PBMC might have exerted a sup
45
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Fig. 4. Percentage of CD16+ cells in PBMC of young and elderly people. The difference between the values of the young and elderly was not statistically significant (p > 0.05).
C'
Percentage o f Monocytes in PBMC
In the next experiment we checked whether PBMC of the elderly were more con taminated with monocytes than those of young persons. It appeared that monocytes constitute about 5-26% of all mononuclear cells. The percentage of monocytes in PBMC of old participants was not statistically higher than that of young subjects (12.7 ± 7.5 in young; 11.8 ± 9.2 in the elderly). Thus, though the content of monocytes in PBMC suspensions of the young and elderly was similar, their suppressive effect on NK activity was different. NK Activity ofPBL
If monocytes in PBMC were the exclusive cells suppressing NK activity, then their re moval should augment the values of NK ac-
46
Fig. 6. NK activity of PBL of young (n = 30) and elderly (n = 35) individuals measured in a 5lCr release assay. The difference between the values of the young and elderly was not significant statistically (p > 0.05).
tivity of the elderly to the level characteristic of young people. To test this, a 5lCr release assay was done using PBL of the same groups whose PBMC were examined earlier (fig. 1). The mode of calculation of the values of NK activity was identical to that used in PBMC tests, i.e. was measured in pairs and expressed as index. The cytotoxic activity of PBL of the elderly appeared to be in the same range as the values of young people; their index was equal to 0.9 (fig. 6). So, monocytes appeared to be the only cells involved in the suppressive effect on NK cell cytotoxic function. They were responsible for the drop of NK activitiy in the elderly, mea surable in a 5lCr release assay. Effect o f Monocytes on N K Activity o f High and Low NK Responders
An analysis of the values of the 5lCr release assay allowed us to detect the low and high responders inside each age group. (Percent cytotoxicity of 20.0 was assumed as a discrim inating value between low and high NK re sponders.) Low NK responders constituted 67.7% of the group of old individuals; among
Mysliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerstcr
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
pressive effect on NK activity of the elderly. In the next experiments we looked for cells which could be responsible for this suppres sive effect. NK activity of the elderly was measured in PBMC depleted of B or T cells, or of monocytes. The results of the depletion experiments (fig. 5) revealed that removal of B cells (CD22+) and T cells carrying CD4 antigen did not have a statistically significant effect on NK cell function of the elderly. Depletion of cells bearing the CD 16 and CD8 antigens, both of which are expressed on the surface of NK cells [14], led to a statistically significant decrease in NK activity in the case of CD8+ cells (p < 0.05) and to an almost complete abolishment of cytotoxic activity in the case of CD 16+cells (p < 0.0001). On the contrary', removal of monocytes produced a statistical ly significant increase in NK activity (p < 6.001 ). These results show that it was the mono cytes that were the only cells hampering NK activity.
125-,
Low responders
High responders
Fig. 7. Effect of depletion of monocytes from PBMC on NK activity of the low ( | ) and high ( 0 ) NK responders among both the young and elderly. The index of increase was calculated as follows: (NK activ ity after depletion/NK activity before depletion) - 1.0. Depletion of monocytes produced a statistically signif icant increase in NK activity in both young (p < 0 .01 ) and elderly (p < 0.001 ) individuals.
Fig. 8. NK activity of elderly subjects measured in a 5lCr release assay with PBL and PBMC containing different concentrations of monocytes. The differences between the values of NK activity in PBL ( | ) and PBMC containing 2% ( ® , 10% (gg) and 15% monocytes were statistically significant (p < 0.05) in the case of the low responders.
young persons they were less numerous, amounting to 44%. Removal of monocytes from PBMC of the elderly caused only a minimal, nonsignificant (p > 0.1) increase in NK activity in the high responders and a substantial, statistically sig nificant (p < 0.001) increase in the low responders. NK activity of the young low responders rose after depletion of monocytes (p < 0.01), while that of the high responders did not change (p > 0.1 ; fig. 7). To confirm the suppressive role of mono cytes on NK activity of the low responders, in the next experiment NK activity of PBL was compared with that of PBMC containing 2. 10 and 15% monocytes. As depicted in figure 8, a concentration of as little as 2% monocytes in PBMC resulted in a statistically significant decrease (p < 0.05) in NK activity in the case of the low NK responders. A similar range of decrease was attained when monocytes constituted 10 and 15% of the total cell suspension. In contrast.
NK activity of the high NK responders did not drop when the concentration of mono cytes in PBMC was 2. 10 and 15%.
The debate about the character of the agerelated changes in the magnitude of NK activ ity is being continued and still the results seem not to be conclusive. Different groups of authors claim quite opposite opinions as to the direction of these changes, i.e. decline [7, 13, 19], increase [6, 9. 10] and lack of changes in NK activity [2. 15. 18] have been ob served. This variability is usually explained by the fact that various authors compare different groups as regards the age and health status. In addition, though the conditions of the cyto toxic test have been standardized, two kinds of effector cells have been used - PBL and PBMC.
47
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Discussion
48
a drop of NK activity after addition of as little as 2% of monocytes, and NK activity of old people became more depressed than that of young ones. Monocytes can profoundly affect NK ac tivity and their effect may be augmentative as well as depressive [5]. While monocyte-de rived prostaglandins are regarded to be sup pressive mediators [4, 5, 8], interleukin-1 [4, 5] and interferons [8] are NK stimulators. The suppressive effect of monocytes on NK activity of the elderly seems to be a simi lar phenomenon as that described in mice. Blair et al. [3] have found that the age-related decline of NK activity in mice can in part be explained by an increase in adherent cell sup pressor function. The results of this paper clearly indicate the importance of the utilization of standard effector cells in the 5lCr release assay. An agerelated decline of NK function was demon strated when PBMC but not PBL were used as effector cells. Although in our study the values of NK activity of the young and elderly became even when PBL were used instead of PBMC. this tendency was not found in the studies of other authors. So, NK activity of PBL of the elderly has been found to be elevated [6,9], decreased [7, 13, 19] or not changed [ 12], The most rele vant differences encountered in these papers concern the age ranges of the elderly individu als. their health status [6, 7, 9, 12, 13, 19] as well as the utilization of the different targets f 13], These factors, then, have to be selected carefully. Application of the anti-CD 16 MoAb as a marker of the cytoiytically active NK cells was based on the finding that CD16-antigen-bearing cells are the most efficient NK effectors [11], The number of these cells has been shown to correlate precisely with NK activity in the elderly [10, 12], in contrast to the lack of such a correlation between NK activity and
Mysliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerster
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
In this paper, we analyzed the effect of the presence of monocytes on the outcome of the NK cytotoxic test in two age groups. To avoid a variability in values due to methodological factors [9] the results of the chromium release assays were assessed in each experiment sepa rately in pairs of young and old donors and the values of young persons were assumed as index = 1.0. The results of the elderly were cal culated in relation to the values of the young subjects. Our results revealed that the mean NK activity of the people aged more than 70 years, measured in PBMC, was lower in com parison with that of the young adults. At the same time the number of cytoiytically active cells in PBMC as well as the cytotoxic poten tial of CD16+ cells, measured in a single-cell cytotoxicity test, did not differ between the two age groups. The same was true for the number of cells able to form conjugates with K562 cells as well as for the cellular composi tion of the conjugates, i.e. percentage of con jugating CD3+, CD4+, CD8+ and CD 16+ cells. These results, together with those docu menting the enhancing effect of the removal of monocytes on NK activity, implied that monocytes present in the suspension of PBMC were responsible for the suppressive effect on NK activity of the elderly. This assumption was verified and confirmed in experiments in which PBL were used as effec tor cells. In this case, the difference between the young and elderly leveled off. The lack of a corresponding diminution of NK activity of PBMC of the elderly when checked in a sin gle-cell cytotoxicity test indicates that the sup pression is probably triggered by cell-to-cell contact. When both age groups of people were divided into individuals exhibiting low and high NK activity, it appeared that low re sponders constituted about 44% of the young group and as much as 67.7% of the old one. Only PBL of low NK responders reacted with
the number of cells bearing the NKH-1 mem brane marker [20], As a conclusion we recommend that both PBL and PBMC should be used as effector cells in order to characterize age-related
changes in NK activity. While NK activity measured in PBMC most probably reflects that existing in vivo, NK activity of PBL cor responds to the actual potential of NK cells.
References 8 Koren. H.S.: Anderson. S.J.. Fisch er. D.G.: Copeland. C.S.; Jensen. P.J.: Regulation o f human natural killing. I. The role o f monocytes, interferon, and prostaglandins. J. Immunol. 127: 2 0 0 7 -2 0 13 (: 981). 9 Krishnaraj. R.: Blandford. G.: Ageassociated alterations in human nat ural killer cells. I. Increased activity as per conventional and kinetic analysis. Cell. Immunol. Immunopathol. 45 : 268-285 (1987). 10 Krishnaraj. R.: Blandford. G.: Ageassociated alterations in human nat ural killer cells. 2. Increased fre quency o f selective NK subsets. Cell. Immunol. 114: 137-148 (1988) . 11 Lanier, L.L.; Le, A.M.; Phillips. J.H.: Warner, N.L.; Babcock. G.F.: Subpopulations o f human natural killer cells defined by expression o f the Leu-7 (HNK-1) and Leu-11 (NK -15) antigens. J. Immunol. 131: 1789-1793(1983). 12 Ligthart. G.J.; Schuit, H.R.: Hijmans. W.: Natural killer cell func tion is not diminished in thehealthy aged and is proportional to the num ber o f NK cells in the peripheral blood. Immunology 68: 396-402 (1989) .
Autoimmunity and normal immune
13 Myiliwska, J.: \1 ysliwski. A.; Witkosvski. J.: Age-dependent decline of natural killer and antibody-depen dent cell-mediated cytotoxicity of human lymphocytes is connected with decrease of their acid phospha tase activity. Mech. Age. Dev . 31 : 1-
functions in aged humans. Arch. Gerontol. Geriatr. 4: 261-271 (1985). 7 Gamazzo. G.: Mirone, G.; Turturici. A.; Favetta. A.; Campo. M.E.; Cosenza. C.: Chiarenza, G.; Stivala. F.: Pathophysiology o f the immune system in the elderly subjects with or without diabetes and variations af ter recombinant interleukin 2. Arch. Gerontol. Geriatr. 9: 163-180 (1989).
15 Press. H.F.: Baines. M.G.: Studies o f natural killer cells. I. In vivo pa rameters affecting normal cytotoxic function. Int. J. Cancer 29: 383-390 (1982). 16 Proust, J.J.: Bender, B.S.: Nagel, J.E.: Adler. W.H.: Developmental biology and senescence; in Nelson, Natural Immunity, pp. 392-439 (Academic Press, New York, 1989). 17 Targan. S.: Dorey. F.: Interferon ac tivation o f ‘pre-spontaneous killer' (pre-SK) cells and alteration in ki netics o f lysis o f both 'pre-SK' and active SK cells. J. Immunol. 124: 2157-2161 (1980). 18 Thompson. J.S.: Wekstein. D.R.: Rhoades, J.L.; Kirkpatrick. C.; Brown. S.A.: Roszman. T.; Straus, R.: Tietz. N.: The immune status o f healthy centenarians. J. Am. Ger iatr. Soc. 2: 274-281 (1984). 19 Tsang, K.Y.; Pan. J.F.; Swanger, D.L..; Fudenberg, H.H.: In vitro res toration o f immune responses in aging humans by isopronosine. Int. J. Immunopharmacol. 7: 199-206 (1985). 20 Vranes, Z.; Uzarevic, B.; Batnic, D.: Peric. M.: Poljakovic, Z.: Marusic. M.: Natural killer (N K H -D ) cell number and activity in health and disease. Med. Lab. Sei. 47: 108-112 (1990).
1 1 (1985).’ 14 Ortaldo, J.R.; Mathieson. B.J.: Wiltrout, R.H.: Characterization and functions o f natural killer cells. Annls Inst. Pasteur, Immunol. 139: 444-450(1988).
49
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
1 Anegon. I; Cuturi. M.C.; Trinchieri, G.; Pcrrusia. G.: Interaction o f Fc receptor (CD 16) ligands induces transcription of interleukin 2 recep tor (CD25) and lymphokine genes, and expression o f their products in human natural killer cells. J. exp. Med. 167: 452-472 (1988). 2 Bender. B.S.: Chrest. F.J.: Adler, W.H.: Phenotypic expression o f nat ural killer cell associated membrane antigens and cytolytic function of peripheral blood cells from different aged humans. J. clin. Lab. Immunol. 21: 3 1-36 (1986). 3 Blair. P.B.: Staskavic. M.O.: Sam. J.S.: Suppression o f natural killer cell activity in voung and old mice. Mech. Age! Dev. 40 : 5 7 -7 0 (19S7). 4 Bloom. E.T.; Babbitt. J.T.: Kawakami. K.: Moncytc-mediated aug mentation o f human natural killer cell activity: Conditions, monocyte and effector cell characteristics. J. Immunol. 137: 172-178(1986). 5 Bloom, E.T.; Babbit, J.T.: Prosta glandin E2, monocyte adherence and interleukin I in the regulation of human natural killer cell activity by monocytes. Nat. Imntun. Cell Growth Regul. 9: 3 6-48 ( 1990). 6 Baton', G.: Szondy. E.: Falus. A.:
Jolanta Mysliwskaa Andrzej Mysliwskia P/o/r Romanowskia /aci'A' Bigdaa Danuta Sosnowskaa 7(?rry Foersterb
Monocytes Are Responsible for Depressed Natural Killer (NK) Activity in both Young and Elderly Low NK Responders
3 Department of Histology, Medical School. Gdansk, and b Geriatric Center. Gdynia-Witomino. Poland
Key Words Human NK activity Age-related differences Monocytes
Abstract Two age groups - young (19-35) and elderly (70-91) - were compared with respect to natural killer (NK) cytotoxic activi ty. In both groups, low and high NK responders could be dis tinguished. Low NK responders constituted about 70% of all elderly and 40% of young individuals. The differences in the magnitude of NK activity among the young and elderly groups could only be observed when peripheral-blood mononuclear cells but not peripheral-blood lymphocytes were used as effec tor cells in a 51Cr release assay. Experiments with removal or addition of graded numbers of monocytes showed that these cells were responsible for the low level of NK activity in both the young and elderly low NK responders.
The ageing process is claimed to be associ ated with a progressive decline of T- and Bcell-dependent immune responses [16]. Stud ies of age-associated alterations in the expres sion of natural killer (NK)-cell-mediated cytolysis revealed all possible changes, i.e. no sig nificant change in the intensity of NK activity [2,15, 18], or increase [6,9, 10] or decrease [7, 13, 19] in the magnitude of response of NK cells.
The age-related increase in NK activity [9] as well as the expansion of the NK cell com partment found in old. healthy individuals [10, 12] may lead to the conclusion that the well-preserved NK cell system in the elderly plays a compensatory role for a decreased function of the T and B cell compartments. The NK cell compartment may then be im portant for survival.
This work was supported in part by the II Polish-AmericanFundM.Sklodowska-Curic.
Dr. J. Myskwska Department of Histology Medical School 7 0 -2 10 Gdansk (Poland)
Received: June 16.1991 Accepted: August 23. 1991
© 1992 S. Kargcr AG, Basel 0304-324X/92/ 0382-0041$2.75/0
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Introduction
Subjects and Methods A group of 35 old volunteers (age range 70-91: 6 men, 29 women), inhabitants of the Old People’s Home in Gdynia-Witomino, was examined. They were under constant medical supervision and regularly un derwent twice a year a general medical examination. An analysis of the following blood parameters was rou tinely done: erythrocyte sedimentation rate: hemoglo bin level; mean corpuscular volume: leukocyte count with differentiation: levels of glucose, urea, lipids, aspartate aminotransferase and alanine aminotransfer ase, and protein electrophoresis. Urinalysis comprised protein, glucose, ketones and sediment. All people admitted to the study were not suffering from chronic infections, malignant disease, lymphoproliférative dis orders and dementia. Only those volunteers were qual ified who had their last acute infection at least 2 months before examination and had not been hospital ized for the last 6 months. Volunteers taking regularly medications, except for vitamins, were not admitted. The control group consisted of 30 young adults medical students and staff members (age range 19-35; 5 men. 25 women) - who were qualified after a careful medical examination. The same blood and urine pa rameters, as in the case of the old group were deter mined. Young volunteers, like elderly, were not suffer ing from acute and chronic infections, malignant dis ease and lymphoproliférative disorders.
42
Preparation o f PBMC Human PBMC were obtained by centrifugation of heparinized blood samples on Lymphoprep gradient (Nyegaard, Norway). After 3 washings with phosphatebuffered saline (PBS), the cells were suspended in RPMI 1640 medium (Gibco. UK) supplemented with 5% fetal calf serum (FCS: Flow Labs.. USA). A me dium composed in such a way will be referred to as ‘medium’. In order to obtain PBL. PBMC samples were incu bated on plastic Petri dishes for 1 h in a humidified atmosphere containing 5% CCH at 37 °C. The plasticnonadherent lymphocytes were separated from plasticadherent monocytes. Different concentrations of monocytes in mononu clear cell suspension were obtained by dilution of PBMC with PBL suspension. Monocytes were identi fied on the basis of the pattern of their acid a-naphthvlesterase (ANAE) activity. B. T and NK cells were depleted from PBL by C'mediated lysis. PBL were incubated for 30 min on ice with one of the following monoclonal antibodies (MoAB): BMA 0130 (anti-CD 19): BMA 081 (antiCD8): BMA 041 (anti-CD4). or BMA 070 (anti-CD 16: Bchringwerke, FRG). After incubation cells were cen trifuged, washed with PBS and incubated with an appropriate amount of guinea pig serum at 37 °C for 30 min. Depletion of a given population was checked with an immunoperoxidase method. In control experi ments, guinea pig serum alone or the irrelevant MoAb BMA 0220 (anti-Reed-Stcrnberg cells: Bchringwerke) was incubated with effector cells which were processed identically as those treated with the relevant MoAb. Preparation o f Target K562 Cells The K.562 human erythroleukemia cell line was maintained in RPMI 1640 medium (Gibco) contain ing 10% FCS (Flow Labs.). 100 pg/ml streptomycin, 100 lU/ml penicillin and 2 m.V//.-glutamine (Gibco). Assessment oJ'NK Cell Activity Single-Cell Cytotoxicity Assay. Two million of both effector (PBMC. PBL) and target (K562) cells sus pended in medium were mixed, centrifuged at 400 g, incubated for 15 min at 37 °C and resuspended. In some experiments, CD16+ cells were marked in the suspension of effector cells before mixing them with K562 cells. PBMC were incubated for 30 min on ice with an appropriate amount of BMA/070 MoAb (anti-CD 16), centrifuged and washed 3 times in PBS. After resuspension cells were incubated for 15 min with an appropriate amount of magnetic Dvnabeads M450 coated with goat antimouse IgG (Dvnal. Nor
MySliwska/MySliwski/Romanowski/ Bigda/Sosnowska/Focrstcr
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
For the determination of NK cytotoxic ac tivity, a 51Cr release assay with K562 cells as targets is routinely applied. As effector cells, some authors use peripheral-blood lympho cytes (PBL) [6, 13] whereas others utilize pe ripheral-blood mononuclear cells (PBMC) [2, 18], The fact that under different experimen tal conditions monocytes can exert various effects on the cytotoxicity of NK cells [5, 8] implies that their presence in the cytotoxic test could be of relevance for the differences in NK activity between young and old per sons. In the present study, we compare NK activity in young (19-35 years old) and el derly (70-91 years old) people. Special atten tion is paid to the differences in NK activity due to the utilization of PBMC and PBL as effector cells.
Staining o f CD3*, CD4*. CD8* and CD 16* cells by Immunoperoxidase Method Conjugates were formed between effector and tar get cells and transferred onto the glass slides. Slides were fixed with acetone. To detect a given cell popula tion, slides were coated with MoAb BMA 030 (antiCD3; diluted 1:5 in PBS/BSA), BMA 040 (anti-CD4), MBA 081 (anti-CD8) or BMA 070 (anti-CD 16). After 30 min of incubation, slides were rinsed with PBS and finally coated for another 30 min with peroxidase-con jugated rabbit antimouse Ig serum (Dako, Denmark, diluted 1:100). After rinsing with PBS. the peroxidase reaction was carried out. The incubation mixture con sisted of 6 mg diaminobenzidine (Sigma, USA) dis solved in 10 ml, of PBS and 5 pi of 30% HiCK The reaction was stopped with distilled water. Slides were counterstained with hematoxylin and finally im mersed with glycerogei. Activity of endogenous peroxidase was abolished by incubation of slides with 1% H2O 2 in methanol for 15 min. Statistical analysis was performed by means of a paired t test.
Results NK Activity o f PBMC o f Elderly Persons
In each experiment. NK activity was deter mined in pairs of young and old donors and the arithmetic mean of the values of the young persons was assumed as index = 1.0. The val ues of 51Cr release of PBMC of the elderly were calculated in relation to the values of the young subjects. As shown in figure 1, the index expressing NK activity of elderly people was 0.543 ± 0.205, which indicates that the intensity of their NK activity was about half that charac teristic of young people. The difference was statistically significant (p < 0.001). Cellular Composition o f Conjugates o f Old and Young Persons
Besides cytolytically active NK effectors characterized by the presence of CD 16 anti gen [11, 14], representatives of the T cell lineage not exhibiting cytotoxic activity be long also to the cells forming conjugates with NK-sensitive targets [12], It is thus possible that conjugates of the elderly were enriched in the nonlytic lymphocytes. To check whether this was really the case, an analysis of the cells in the conjugates was performed. It was found that about 98% of conjugates consisted of one target and one effector cell. The percentage of cells forming conjugates was similar in both age groups: 15.4 ± 1.3% in the young and 14.8 ± 3.2% in the elderly. There were no sta tistically significant differences between the participation of conjugating cells carrying CD 16. CD3, CD4 and CD8 antigens (fig. 2). As follows from our results. PBMC of old individuals formed as many conjugates with K562 targets as those of young subjects. Moreover, conjugates of old and young indi viduals were not distinctly composed with respect to the cells expected to be both lytic and nonlytic.
43
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
way). The suspension of effector and target cells was gently mixed with an equal volume of warm (3942 °C). liquified agarose (Windsor Labs., UK) to give a final concentration of agarose of 0.5%. The mixture was then layered on plastic Petri dishes (Technomcd, Poland). After solidification of agarose, medium was poured on the agarose surface. The cells were incu bated for 3 h at 37 °C in a humidified atmosphere con taining 5% COi. After incubation dishes were stained with 1% solution of trypan blue. The percentage of tar get-binding cells with dead targets was counted. The cytotoxic activity of cells in conjugates (cytotoxic in dex; Cl) was calculated according to the formula of Targan and Dorey [ 17): Cl = percentage of dead targets in conjugates - (fraction of spontaneous dead targets X percentage of dead targets in conjugates). 5lCr Release Assay. The samples, run in triplicate, contained a mixture of 2 X 104 K562 cellsand 5 X I05 effector cells. The volume of samples was adjusted to 0.4 ml. The samples were incubated in 5% COi at 37°C for 4h. then centrifuged at 2,000 rpm for 10 min. The 0.25-ml supernatants were removed and counted for gamma radioactivity in an LKB gamma counter. The specific cytotoxicity was computed by the following formula: percent chromium release = (exper imental release - spontaneous release)/(maximal re lease - spontaneous release).
1 . 2-1
Young
Elderly
Fig. 1. NK activity of PBMC of young (n = 30) and elderly (n = 35) individuals measured in a 51Cr release assay. In each experiment NK activity was determined in pairs of young and old donors, and the arithmetic mean of the values of young persons was assumed as index = 1.0. Values of the elderly were calculated in relation to the values of the young subjects. The differ ence between the values of the young and elderly was statistically significant (p < 0.001).
Fig. 2. Percentage of conjugates formed by CD 16*. CD3*. CD4‘ and CD8* cells in young (ED and elderly ( B individuals. The differences between the values of the young and elderly in each conjugate group were not statistically significant (p > 0.05).
Cytotoxic Activity o f CD I6+Cells Expressed at a Single-Cell Level
44
Fig. 3. Cytotoxic activity as assessed in a single-cell cytotoxicity assay with PBMC, and cells bearing (CD16*) and lacking (CDI6-) CD16 antigen. The dif ferences between the activity of CDI6+ and CD 16populations were statistically highly significant (p < 0.0001). Young subjects: 0 « elderly subjects.
values for the two age groups, pointing that cells not carrying the CD 16 marker were also not responsible for the differences in NK activity between young and old participants either. Since the anti-CD 16 MoAb used as a marker of NK cells in a single-cell cytotoxic ity test could have an impact on NK activity [1], in the next experiment a single-cell cyto-
MySliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerster
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Thus, although CD16+ cells of the elderly were contained in conjugates in a similar pro portion to those of the young subjects, the cytotoxic activity of the elderly was reduced. This fact implied that some CD 16+cells of the elderly might have been devoid of cytolytic activity. This possibility was tested in the next experiment in which CD 16-antigen-bearing cells, regarded as the strongest cytolytically active NK population [11], were marked by an immunomagnetic method and tested in a single-cell cytotoxicity test in agarose. As shown in figure 3. almost all NK activ ity of both age groups resided in the subpopu lation of CD16+ cells, confirming that these cells were the actually cytolytically active ones. CD 16+cells of the elderly people were as active as those of the young group (p > 0.05). The Cl of the CD 16" population, though markedly lower, was also in a similar range of
2.0-,
-C D 22
¡MoAb -CD4 -C D 8 -C D 16 PBMC
PBL
Fig. 5. Effect of depletion of different cell populations from PBMC on NK activity of elderly people as measured in a 5lCr release assay. NK activity is shown after depiction of CD22+ cells (-CD22), in PBMC incu bated with C', without MoAb (C') in PBMC incubated with an irrelevant MoAB (iMoAb) after depletion of CD4' cells (-CD4). CD8* cells (-CD8) and CD16- cells (-CD16). in nontreated PBMC and in PBL (PBL). The index of increase was calculated as follows: (NK activity after dcpletion/NK activity before depletion) - l.O. Depletion of monocytes pro duced a statistically significant increase (p < 0.001) in NK activity.
toxicity test was carried out on the nonmarked PBMC and again the Cl of both groups was compared. Cl of the young was 21 ± 5.6 and that of the elderly 19.5 ± 7.1. This indicated that anti-CD 16 MoAb did not af fect the outcome of a single-cell cytotoxicity test and reinforced the former finding that the cytolytic effectiveness of all cytolytically ac tive cells was equal in both age groups. Taken together, these results demonstrate that the defect of NK activity observed in the group of elderly people, measurable in a 5lCr release assay, does not depend on the reduced number of active NK cells. Percentage o f CD 16* Cells in Blood As a reduced number of cytotoxically ac tive NK cells in conjugates did not account for the lower cytotoxic activity of PBMC of
the elderly, it was possible that elderly people had less CD16+ cells in blood and therefore a reduced general cytotoxic potential of their PBMC, as reflected in a 5lCr release assay. This possibility was tested in the next experi ment. It appeared, however, that both age groups possessed a comparable percentage of CD 16+ cells in the blood (fig. 4). Effect o f Depletion o f Different Celt Populations on NK Activity
The discrepancy between the tendency of the NK activity values of old people to drop when measured in the 5lCr release assay and to be on the same level as those of the young when assessed in a single-cell cytotoxicity as say suggests that under the conditions of the 51Cr release assay a certain population of cells present in PBMC might have exerted a sup
45
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Fig. 4. Percentage of CD16+ cells in PBMC of young and elderly people. The difference between the values of the young and elderly was not statistically significant (p > 0.05).
C'
Percentage o f Monocytes in PBMC
In the next experiment we checked whether PBMC of the elderly were more con taminated with monocytes than those of young persons. It appeared that monocytes constitute about 5-26% of all mononuclear cells. The percentage of monocytes in PBMC of old participants was not statistically higher than that of young subjects (12.7 ± 7.5 in young; 11.8 ± 9.2 in the elderly). Thus, though the content of monocytes in PBMC suspensions of the young and elderly was similar, their suppressive effect on NK activity was different. NK Activity ofPBL
If monocytes in PBMC were the exclusive cells suppressing NK activity, then their re moval should augment the values of NK ac-
46
Fig. 6. NK activity of PBL of young (n = 30) and elderly (n = 35) individuals measured in a 5lCr release assay. The difference between the values of the young and elderly was not significant statistically (p > 0.05).
tivity of the elderly to the level characteristic of young people. To test this, a 5lCr release assay was done using PBL of the same groups whose PBMC were examined earlier (fig. 1). The mode of calculation of the values of NK activity was identical to that used in PBMC tests, i.e. was measured in pairs and expressed as index. The cytotoxic activity of PBL of the elderly appeared to be in the same range as the values of young people; their index was equal to 0.9 (fig. 6). So, monocytes appeared to be the only cells involved in the suppressive effect on NK cell cytotoxic function. They were responsible for the drop of NK activitiy in the elderly, mea surable in a 5lCr release assay. Effect o f Monocytes on N K Activity o f High and Low NK Responders
An analysis of the values of the 5lCr release assay allowed us to detect the low and high responders inside each age group. (Percent cytotoxicity of 20.0 was assumed as a discrim inating value between low and high NK re sponders.) Low NK responders constituted 67.7% of the group of old individuals; among
Mysliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerstcr
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
pressive effect on NK activity of the elderly. In the next experiments we looked for cells which could be responsible for this suppres sive effect. NK activity of the elderly was measured in PBMC depleted of B or T cells, or of monocytes. The results of the depletion experiments (fig. 5) revealed that removal of B cells (CD22+) and T cells carrying CD4 antigen did not have a statistically significant effect on NK cell function of the elderly. Depletion of cells bearing the CD 16 and CD8 antigens, both of which are expressed on the surface of NK cells [14], led to a statistically significant decrease in NK activity in the case of CD8+ cells (p < 0.05) and to an almost complete abolishment of cytotoxic activity in the case of CD 16+cells (p < 0.0001). On the contrary', removal of monocytes produced a statistical ly significant increase in NK activity (p < 6.001 ). These results show that it was the mono cytes that were the only cells hampering NK activity.
125-,
Low responders
High responders
Fig. 7. Effect of depletion of monocytes from PBMC on NK activity of the low ( | ) and high ( 0 ) NK responders among both the young and elderly. The index of increase was calculated as follows: (NK activ ity after depletion/NK activity before depletion) - 1.0. Depletion of monocytes produced a statistically signif icant increase in NK activity in both young (p < 0 .01 ) and elderly (p < 0.001 ) individuals.
Fig. 8. NK activity of elderly subjects measured in a 5lCr release assay with PBL and PBMC containing different concentrations of monocytes. The differences between the values of NK activity in PBL ( | ) and PBMC containing 2% ( ® , 10% (gg) and 15% monocytes were statistically significant (p < 0.05) in the case of the low responders.
young persons they were less numerous, amounting to 44%. Removal of monocytes from PBMC of the elderly caused only a minimal, nonsignificant (p > 0.1) increase in NK activity in the high responders and a substantial, statistically sig nificant (p < 0.001) increase in the low responders. NK activity of the young low responders rose after depletion of monocytes (p < 0.01), while that of the high responders did not change (p > 0.1 ; fig. 7). To confirm the suppressive role of mono cytes on NK activity of the low responders, in the next experiment NK activity of PBL was compared with that of PBMC containing 2. 10 and 15% monocytes. As depicted in figure 8, a concentration of as little as 2% monocytes in PBMC resulted in a statistically significant decrease (p < 0.05) in NK activity in the case of the low NK responders. A similar range of decrease was attained when monocytes constituted 10 and 15% of the total cell suspension. In contrast.
NK activity of the high NK responders did not drop when the concentration of mono cytes in PBMC was 2. 10 and 15%.
The debate about the character of the agerelated changes in the magnitude of NK activ ity is being continued and still the results seem not to be conclusive. Different groups of authors claim quite opposite opinions as to the direction of these changes, i.e. decline [7, 13, 19], increase [6, 9. 10] and lack of changes in NK activity [2. 15. 18] have been ob served. This variability is usually explained by the fact that various authors compare different groups as regards the age and health status. In addition, though the conditions of the cyto toxic test have been standardized, two kinds of effector cells have been used - PBL and PBMC.
47
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
Discussion
48
a drop of NK activity after addition of as little as 2% of monocytes, and NK activity of old people became more depressed than that of young ones. Monocytes can profoundly affect NK ac tivity and their effect may be augmentative as well as depressive [5]. While monocyte-de rived prostaglandins are regarded to be sup pressive mediators [4, 5, 8], interleukin-1 [4, 5] and interferons [8] are NK stimulators. The suppressive effect of monocytes on NK activity of the elderly seems to be a simi lar phenomenon as that described in mice. Blair et al. [3] have found that the age-related decline of NK activity in mice can in part be explained by an increase in adherent cell sup pressor function. The results of this paper clearly indicate the importance of the utilization of standard effector cells in the 5lCr release assay. An agerelated decline of NK function was demon strated when PBMC but not PBL were used as effector cells. Although in our study the values of NK activity of the young and elderly became even when PBL were used instead of PBMC. this tendency was not found in the studies of other authors. So, NK activity of PBL of the elderly has been found to be elevated [6,9], decreased [7, 13, 19] or not changed [ 12], The most rele vant differences encountered in these papers concern the age ranges of the elderly individu als. their health status [6, 7, 9, 12, 13, 19] as well as the utilization of the different targets f 13], These factors, then, have to be selected carefully. Application of the anti-CD 16 MoAb as a marker of the cytoiytically active NK cells was based on the finding that CD16-antigen-bearing cells are the most efficient NK effectors [11], The number of these cells has been shown to correlate precisely with NK activity in the elderly [10, 12], in contrast to the lack of such a correlation between NK activity and
Mysliwska/Mysliwski/Romanowski/ Bigda/Sosnowska/Foerster
Human NK Activity
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
In this paper, we analyzed the effect of the presence of monocytes on the outcome of the NK cytotoxic test in two age groups. To avoid a variability in values due to methodological factors [9] the results of the chromium release assays were assessed in each experiment sepa rately in pairs of young and old donors and the values of young persons were assumed as index = 1.0. The results of the elderly were cal culated in relation to the values of the young subjects. Our results revealed that the mean NK activity of the people aged more than 70 years, measured in PBMC, was lower in com parison with that of the young adults. At the same time the number of cytoiytically active cells in PBMC as well as the cytotoxic poten tial of CD16+ cells, measured in a single-cell cytotoxicity test, did not differ between the two age groups. The same was true for the number of cells able to form conjugates with K562 cells as well as for the cellular composi tion of the conjugates, i.e. percentage of con jugating CD3+, CD4+, CD8+ and CD 16+ cells. These results, together with those docu menting the enhancing effect of the removal of monocytes on NK activity, implied that monocytes present in the suspension of PBMC were responsible for the suppressive effect on NK activity of the elderly. This assumption was verified and confirmed in experiments in which PBL were used as effec tor cells. In this case, the difference between the young and elderly leveled off. The lack of a corresponding diminution of NK activity of PBMC of the elderly when checked in a sin gle-cell cytotoxicity test indicates that the sup pression is probably triggered by cell-to-cell contact. When both age groups of people were divided into individuals exhibiting low and high NK activity, it appeared that low re sponders constituted about 44% of the young group and as much as 67.7% of the old one. Only PBL of low NK responders reacted with
the number of cells bearing the NKH-1 mem brane marker [20], As a conclusion we recommend that both PBL and PBMC should be used as effector cells in order to characterize age-related
changes in NK activity. While NK activity measured in PBMC most probably reflects that existing in vivo, NK activity of PBL cor responds to the actual potential of NK cells.
References 8 Koren. H.S.: Anderson. S.J.. Fisch er. D.G.: Copeland. C.S.; Jensen. P.J.: Regulation o f human natural killing. I. The role o f monocytes, interferon, and prostaglandins. J. Immunol. 127: 2 0 0 7 -2 0 13 (: 981). 9 Krishnaraj. R.: Blandford. G.: Ageassociated alterations in human nat ural killer cells. I. Increased activity as per conventional and kinetic analysis. Cell. Immunol. Immunopathol. 45 : 268-285 (1987). 10 Krishnaraj. R.: Blandford. G.: Ageassociated alterations in human nat ural killer cells. 2. Increased fre quency o f selective NK subsets. Cell. Immunol. 114: 137-148 (1988) . 11 Lanier, L.L.; Le, A.M.; Phillips. J.H.: Warner, N.L.; Babcock. G.F.: Subpopulations o f human natural killer cells defined by expression o f the Leu-7 (HNK-1) and Leu-11 (NK -15) antigens. J. Immunol. 131: 1789-1793(1983). 12 Ligthart. G.J.; Schuit, H.R.: Hijmans. W.: Natural killer cell func tion is not diminished in thehealthy aged and is proportional to the num ber o f NK cells in the peripheral blood. Immunology 68: 396-402 (1989) .
Autoimmunity and normal immune
13 Myiliwska, J.: \1 ysliwski. A.; Witkosvski. J.: Age-dependent decline of natural killer and antibody-depen dent cell-mediated cytotoxicity of human lymphocytes is connected with decrease of their acid phospha tase activity. Mech. Age. Dev . 31 : 1-
functions in aged humans. Arch. Gerontol. Geriatr. 4: 261-271 (1985). 7 Gamazzo. G.: Mirone, G.; Turturici. A.; Favetta. A.; Campo. M.E.; Cosenza. C.: Chiarenza, G.; Stivala. F.: Pathophysiology o f the immune system in the elderly subjects with or without diabetes and variations af ter recombinant interleukin 2. Arch. Gerontol. Geriatr. 9: 163-180 (1989).
15 Press. H.F.: Baines. M.G.: Studies o f natural killer cells. I. In vivo pa rameters affecting normal cytotoxic function. Int. J. Cancer 29: 383-390 (1982). 16 Proust, J.J.: Bender, B.S.: Nagel, J.E.: Adler. W.H.: Developmental biology and senescence; in Nelson, Natural Immunity, pp. 392-439 (Academic Press, New York, 1989). 17 Targan. S.: Dorey. F.: Interferon ac tivation o f ‘pre-spontaneous killer' (pre-SK) cells and alteration in ki netics o f lysis o f both 'pre-SK' and active SK cells. J. Immunol. 124: 2157-2161 (1980). 18 Thompson. J.S.: Wekstein. D.R.: Rhoades, J.L.; Kirkpatrick. C.; Brown. S.A.: Roszman. T.; Straus, R.: Tietz. N.: The immune status o f healthy centenarians. J. Am. Ger iatr. Soc. 2: 274-281 (1984). 19 Tsang, K.Y.; Pan. J.F.; Swanger, D.L..; Fudenberg, H.H.: In vitro res toration o f immune responses in aging humans by isopronosine. Int. J. Immunopharmacol. 7: 199-206 (1985). 20 Vranes, Z.; Uzarevic, B.; Batnic, D.: Peric. M.: Poljakovic, Z.: Marusic. M.: Natural killer (N K H -D ) cell number and activity in health and disease. Med. Lab. Sei. 47: 108-112 (1990).
1 1 (1985).’ 14 Ortaldo, J.R.; Mathieson. B.J.: Wiltrout, R.H.: Characterization and functions o f natural killer cells. Annls Inst. Pasteur, Immunol. 139: 444-450(1988).
49
Downloaded by: Stockholm University Library 130.237.165.40 - 12/17/2018 10:00:57 PM
1 Anegon. I; Cuturi. M.C.; Trinchieri, G.; Pcrrusia. G.: Interaction o f Fc receptor (CD 16) ligands induces transcription of interleukin 2 recep tor (CD25) and lymphokine genes, and expression o f their products in human natural killer cells. J. exp. Med. 167: 452-472 (1988). 2 Bender. B.S.: Chrest. F.J.: Adler, W.H.: Phenotypic expression o f nat ural killer cell associated membrane antigens and cytolytic function of peripheral blood cells from different aged humans. J. clin. Lab. Immunol. 21: 3 1-36 (1986). 3 Blair. P.B.: Staskavic. M.O.: Sam. J.S.: Suppression o f natural killer cell activity in voung and old mice. Mech. Age! Dev. 40 : 5 7 -7 0 (19S7). 4 Bloom. E.T.; Babbitt. J.T.: Kawakami. K.: Moncytc-mediated aug mentation o f human natural killer cell activity: Conditions, monocyte and effector cell characteristics. J. Immunol. 137: 172-178(1986). 5 Bloom, E.T.; Babbit, J.T.: Prosta glandin E2, monocyte adherence and interleukin I in the regulation of human natural killer cell activity by monocytes. Nat. Imntun. Cell Growth Regul. 9: 3 6-48 ( 1990). 6 Baton', G.: Szondy. E.: Falus. A.:
