- Email: [email protected]
Anti-Candidial Activity of Natural Killer (NK) and Lymphokine Activated Killer (LAK) Lymphocytes In Vitro
Immunobiol., vol. 195, pp. 220-230 (1996)
©
1996 by Gustav Fischer Verlag, Stuttgart
Department of Microbiology, Gazi University, Faculty of Medicine, Be§evler, Ankara, Turkey
Anti-Candidial Activity of Natural Killer (NK) and Lymphokine Activated Killer (LAK) Lymphocytes In Vitro ZEYNEP GOLAY
and TURGUT IMIR
Received December 8, 1994 . Accepted in revised form October 11, 1995
Abstract The natural cytotoxic effects of peripheral blood lymphocytes (PBL) on Candida stellatoidea and several other Candida species were examined by a colony forming inhibition (CFI) assay. Peripheral blood mononuclear cells (PBMC), were incubated with C. stellatoidea yeast cells. After the incubation period the colony-forming ability of the yeast was significantly reduced. In similar experiments, six different Candida species (c. albicans, C. krusei, C. stellatoidea, C. tropicalis, C. pseudotropicalis, C. guillermondii) were used as target cells. There was no statistically significant difference in the anticandidial activities of PBL against the Candida species used. It was demonstrated that a fraction of lymphocytes, natural killer cells (NK), had the major natural anti-c
Abbreviations: NK = natural killer; LAK = lymphokine activated killer; PBL = peripheral blood lymphocytes; CFI = colony forming inhibition; PBMC = peripheral blood mononuclear cells; mAbs = monoclonal antibodies; JL-2 = saline; FCS = foetal calf serum; IFN-y = interferon-y
NK and LAK activity on Candida . 221
Introduction
NK cells, a component of natural cell-mediated CytotOXICity, are a. heterogeneous population of granular lymphoid cells that lyse certain tumor and virally-infected cells without prior sensitization (1). The NK cell is functionally a subset of lymphocytes (2). NK cells are very heterogeneous with regard to cell surface antigen structure, but with the recent development of monoclonal antibodies (mAbs) the analysis of these antigens could be more comprehensive (3, 4). One of the widely used mAbs for this purpose is anti-CDl6 antibody which binds the low affinity IgG Fe III receptor and reacts with essentially all human NK cells in peripheral blood (5). Anti-CD 16 antibodies recognize 5~0 kDa antigen present on human NK cell membrane and is expressed on approximately 15 % of peripheral blood lymphocytes. Cross-linking of CD16 as well as CD45 (leukocyte common antigen) can stimulate cytokine production (6). It was also reported that CD16 antigen might be expressed on CD 3 positive T lymphocytes in certain individuals (7, 8). It is also expressed on neutrophils, eosinophils, monocytes and macrophage subsets (9). CD15 antigen, the lacto-N-fucopentaeose III molecule, is a branched pentasaccharide with a molecular weight of 185-200 kDa. This carbohydrate moiety can be found in glycolipids and glycoproteins present on cell membrane. It is strongly expressed by neutrophils, eosinophils, monocytes and normal myeloid precursor cells (9). IL-2, a lymphokine produced by T lymphocytes upon stimulation with either mitogens or antigens, is a 15.000 MW glycoprotein in human beings and has a variety of immunoregulatory properties. It is a known activator of NK antitumoral activity (10). Lymphokine-activated killer (LAK) cells, capable of killing malignant target cells in a non-MHC restricted manner, are generated from NK cells upon exposure to IL-2 for 72 h (12). NK cytotoxic activity was first defined for tumor cell lines as they have natural receptors for tumour cell lines such as K562 and Cr51 release assay is used to measure NK cytotoxic activity (13). In recent years, the anti-microbial effect of NK cells has attracted a great deal of attention (14). The in vivo (15) and in vitro (16, 17) effect of NK cells on the various forms of C. albicans has been shown by different methods as indicated in previous reports. In the present study, anti-candidial NK and LAK activity by inhibition of colony formation of various Candida species, is described.
222 . Z. GULAY and T. IMIR
Materials and Methods Fungi
Candida species used throughout this study, were isolated from clinical specimens in our laboratory and identified according to established taxonomic criteria (18). The fungi were grown on a blood agar base for 18 h and the colonies collected while they were on exponential growth term. Yeast cells were resuspended in RPMI 1640 medium (Flow Lab. U.K.) and adjusted to the concentration of 8 x 103 viable organisms per ml. Isolation of peripheral blood lymphocytes (PBL)
Heparinized blood from healthy male donors aged between 20 and 40 was layered on Ficoll-Isopaque separation medium (histopaque 1077, Sigma, St. Louis, Mo, USA) and PBL were isolated by density centrifugation (11). PBL were washed in Phosphate Buffered Saline (PBS) at least three times and resuspended in RPMI 1640 medium supplemented with L-glutamine. The suspension of PBL was poured into Petri dishes sensitized with autologous serum and was incubated for 1 h at 37 °C. After incubation, the non-adherent cells were gently harvested and the same procedure was applied second time. The non-adherent cells were washed in RPMI 1640 and used as the effector cells. The relative percentage of monocytes was less than 5 percent as determined by morphological examination following Giemsa staining. The average cell viability was not less than 98 % as determined by trypan blue dye exclusion test. Cell depletion with mAbs
Anti-Leu lIb (anti-CD 16), anti-Leu Ml (anti-CD 15) monoclonal antibodies (Becton Dickinson, Mountain View, Ca, USA) and baby rabbit complement were used. PBL (5 x 10 6 /ml) was separated into eight groups and each group was incubated with the reagents as follows: 1) anti-Leu 11 b + complement, 2) anti-Leu Ml + complement, 3) anti-Leu I1b+Ml + complement, 4) complement,S) anti-Leu l1b, 6) anti-Leu Ml, 7) only RPMI 1640, 8) IL-2 according to the procedure described earlier (5). Cells were then washed and counted, it was found that 13 % of the total mononuclear cells were CD 16+ and 4 % were CD 15+. The absence of any residual cells reacting with the relevant antibodies after complement-mediated lysis was verified by immunofluorescence. The PBLs were counted, their number corrected and used as effectors in the assay. Effect of IFN-y on cytotoxicity
In order to investigate whether the cytotoxic activity on colony inhibition was a reason for IFN-y, we used anti-IFN-y monoclonal antibody (Boehringer Mannheim, Germany) to neutralize if any IFN-y which is secreted mostly by NK cells. After addition of 10 f!g/ml anti-IFN-y to each well, inhibitory effect of NK cells was measured using the anticandidial cytotoxicity assay. Activation with recombinant interleukin-2 (IL-2)
rhL-2 (Sigma, St. Louis, MO, USA) was used in various concentrations from 3 to 100 U I
ml for 106 lymphocytes. Monocyte depleted PBL were incubated with rhL-2 for 72 hand used as lymphokine activated killer (LAK) cells (11). The viability o f PBLs after incubation was always more than 90 % determined by trypan blue dye exclusion.
NK and LAK activity on Candida . 223 Colony forming inhibition assay (CFI) C. stellatoidea yeast cells, used as target cells (unless otherwise stated), were incubated with effector cells in 96-well U-bottomed microtitre plates at different target:effector ratios (0.1 ml of a 8 x 103 yeast cells/ ml Candida suspension with 0.1 ml of effector cell suspensions) for 2 h at 37 °C in a 5 % CO 2 incubator (16). Target cells, incubated in RPMI 1640 medium alone were used as controls. After incubation, aliquots of 25 [.ll from each tube were plated on Petri dishes containing brain heart infusion agar. Cultures were then incubated for 48 h at 37°C, colonies were counted (100 colonies per plate expected on control plates) and the percentage reduction in colony numbers was determined as the Anti-Candidial Index (ACI) according to the following formula:
ACI= ( 1-
experimental colony number) x 100 control colony number
Each experiment was performed in quadruplicate. The results were represented as the mean ± SEM values of ACI obtained from separate groups. Chromium release assay for NK and LAK cell activity
K562 (for the NK assay) and Daudi (for the LAK assay) cells were labelled by incubating 5 x 106 target cells with 0.1 ml Na/1Cr04 (specific activity 1.0 mCi/ ml, Amersham, U.K.) for 1 h at 37"C in a humidified air plus 5 % CO 2, washed three times and then 10 4 target cells in 0.1 ml was used in each micro titre well. Effector cells, at a density of 25 times the target cell number (E:T 25:1) in 0.1 ml volume, were added to labelled cells in the wells. Spontaneous release was estimated by incubating target cells with 0.1 ml medium only. Maximum release was determined by incubating labelled target cells with 0.1 ml of 2 % sodium dodecyl sulphate. After a 4-h incubation period, the microtitre plates were centrifuged at 150 x g for 5 min and 0.1 ml of supernatants from each well was counted in a gamma counter (LKB, 1275 minigamma). Cytotoxicity was calculated from the average counts per minute (cpm) released into the supernatants of triplicate samples using the following formula (5): % specific cytotoxicity =
test cpm - spontaneous cpm x 100 max. cpm - spontaneous cpm
Statistical ana'lysis
The results are expressed as the mean ± SEM. The significant difference between two means was determined by the Student's t-test.
Results Anti-candidial effect of lymphocytes
When freshly isolated and monocyte-depleted human PBL from healthy male donors was incubated for 1, 2 and 4 h with C. stellatoidea, the ability of Candida to form colonies was inhibited. Representative results obtained from four donors are presented in Figure 1. The spontaneous cytotoxicity of PBL against C. stellatoidea from different donors varied somewhat but
224 . Z.
GOLAY
and T.
IMIR
60
1:5
1:10 1:15
1:5
1:10 1:15
1:5
1:10 1:15 T:E
1h
2h
4h
Incubation Time Figure 1. PBL's anti-candidial effects on C. stellatoidea. The standard error of all determinants was less than 7 %. n = 4.
all donors demonstrated anti-candidial activity. The level of inhibition of colony formation (anti-candidial index), was dependent on effector cell number and incubation period. Longer incubation of effector/target mixture resulted in higher inhibition. Target: effector cell ratios of 1:5, 1: 10 and 1:15 were used for ACI. After 1 h incubation period, the mean ± SEM values of % inhibition in colony-forming were 9.5 ± 4.8, 15.5 ± 5.1 and 26.2 ± 6.7 for T:E ratios of 1:5, 1:10 and 1:15, respectively. For 2 h incubation period: 37.2 ± 3.7, 45.9 ± 5.8 and 53.6 ± 5.8 and for 4 h incubation period: 48,8 ± 2.2, 56,0 ± 4,5 and 62.5 ± 6.0 percent inhibition were achieved at the same T:E ratios.
Table 1. The mean values of PBL's ACI for different species. 2 h ACI activity was performed at the T:E ratio of 1:10. Data were expressed as the mean ± SEM, n=4.
C. C. C. C. C. C.
albicans stellatoidea tropicalis pseudotropicalis krusei guillermondii
28.3 ± 3.4"29.9±2.7"" 28.6 ± 1.6':29.4 ± 2.5" 32.2 ± 2.2" 32.1 ± 1.0':-
NK and LAK activity on Candida . 225 Table 2. The effect of anti IFN-y on ACI. The experiment is representative of three performed, 'T:E= 1:10, b mean ± SEM. ACI a 38.4 ± 3.6 b 36.1 ± 4.3
Control (C) C + anti-IFN-y
Different Candida species and ACI activity
ACI of effector cells against 6 different Candida species (c. stellato idea, C. albicans, C. tropicalis, C. pseudotropicalis, C. guillermondii, C. krusei) was compared in 7 different experiments using the same effector cells obtained from one donor. Although some strains are established pathogens (for example C. albicans), it is important to state that all the species used were isolated from clinical specimens. At the target:effector ratio of 1: 10 no statistically significant difference in ACI was noted among different Candida species (Table 1). The effect of anti-IFN-y, anti-Leu 11 b, anti-Leu M 1 and complement pretreatment of PBL an ACI
In this experiment anti-IFN-y was used as a cytokine inhibitor to show that the cytotoxic mechanism is not due to IFN -y secretion. As is known that it has an activating effect on NK cells to lyse target cells, anti-IFN-y was present during 4 h of incubation period. The results demonstrated that there was no measurable inhibition of NK activity (Table 2). Table 3. The effect of monoclonal antibodies and complement on ACI (ACI values are expressed as mean ± SEM, the numbers in parentheses correspond to the % inhibition in ACI compared with control PBL in RPMI 1640 only). The cell viabil ity was> 95 % in all cultures and this did not change with or without exposure to mAbs and/ or complement. Specific cell populations indicated were removed by complement-mediated lysis. C. stellatoidea was used as target cells. Data were represented as the mean ± SEM of three different experiments. ACI
1:5 Control (C) C + Complement (C') anti-Leu 11 b anti-Leu Ml + C' anti-Leu 11 b + C' anti-Leu M1 + anti-11 b + C'
26.7 ± 3.2 27.2 ± 2.5 25.7 ± 6.0 20.6 ± 2.0 3.8 ± 1.4 1.5 ± 2.1
1:15
1:10
(0) (4) (23) (86) (95)
43.5 ± 6.4 45.5 ± 2.8 42.3 ± 4.6 36.5 ± 6.0 9.0 ± 5.4 1.6 ± 3.8
(0) (0) (16) (79) (96)
49.0 ± 6.9 51.5 ± 5.2 (0) 49.5 ± 10.6 (0) 41.5 ± 7.2 (15) 10.7 ± 5.0 (78) 5.0 ± 6.9 (90)
226 . Z.
GULAy
and T. IMIR
Table 4. Enhancement of anti-candidial activity by rhIL-2." Activation relative to untreated control, a PBL incubated with IL-2 for 72 h, b mean values ± SEM, CP < 0.05, the experiment shown is representative of three performed. ACI (T:E) IL-2 Concentration (U/ml/l06 PBL)
o (Control)a 3 10 30 100
1:5 (%)*
1:10 (%)
1:15 (%)
24.1 ± 2.9 b 27.8 ± 6.6 (15.4) 40.2 ± 5.2 (66.8) 43.0 ± 4.8 (78.4)C 42.0 ± 14.8 (74.3)C
31.7 ± 4.4 36.7 ± 2.3 (15.7) 47.6 ± 6.1 (49.9) 52.7 ± 5.8 (66.0)C 54.0 ± 11.2 (70.0t
39.5 ± 2.2 45.8 ± 6.3 (15.9) 55.1 ± 4.0 (39.6) 60.9 ± 7.6 (54.2t 64.4 ± 13.9 (58.4)"
To show which subpopulation of lymphocytes has natural anti-candidial activity, anti-Leu llb, anti-Leu Ml monoclonal antibodies plus complement were used. In the anti-Leu lIb + Leu Ml + C'group, anti-candidial activity was reduced by 96 % (p < 0.01) (Table 3). ACI was not changed either with complement or anti-Leu 11 b group, but strong inhibition was observed when both components were used together. Anti-Leu Ml antibody plus complement diminished ACI insignificantly. Therefore, it was concluded that most natural anti-candidial activity was performed in the NK cell population. Enhancement of anti-candidial activity by rhlL-2
Adherent, cell-depleted PBL from 4 healthy donors was incubated for 72 h with 0 (control), 3,10,30 and 100U/ml rhIL-2, and then assayed for anticandidial activity. As seen in Table 4, effect of rhIL-2 was dose-dependent, the magnitude of response increased proportionally with the higher doses of rhIL-2. Three days' incubation of PBL with rhIL-2, strongly stimulated Table 5. Inhibition of NK activity on ACI and cytotoxicity by mAb + c. aT:E = 1:5, bT:E= 1:25, cPBL incubated with and without 100 Vlml IL-2 for 72 h, d mean ± SEM, e PBL treated with mAb + C, as described in Materials and Methods. The representative of 4 experiments is shown.
Target cells
ACP
% Cytotoxicityb
C. stellatoidea
K562
HIL2 Control PBL PBL + anti-Leu 11 b + C e
(+) IL2 C
HIL2
Daudi (+) IL2
HIL2
(+) IL2
15.2 ± 3.5d 27.6 ± 4.9 21.8 ± 5.2 37.3 ± 6.4 12.3 ± 4.4 34.2 ± 5.8 5.8 ± 4.1
6.5 ± 3.2
8.4 ± 3.7
9.8 ± 5.7
7.4 ± 3.3 11.7 ± 5.6
NK and LAK activity on Candida . 227
the natural anti-candidial activity. Adherent cell-depleted effector cells were pre-treated with anti-Leu llb, anti-Leu Ml and complement as described in Materials and Methods prior to stimulation with rhIL-2. As seen in Table 5, the NK activity on C stellatoidea was significantly reduced after treatment with mAbs plus complement and the enhancing activity of rhIL-2 was nearly abolished. Effector cells were incubated with 100 V/ml rhIL-2 for 72 h before the ACI assay.
Discussion A subpopulation of human peripheral blood lymphocytes from healthy donors are able to lyse a variety of target cells, including tumor and virusinfected cells, without prior sensitisation. The effector cells responsible for this spontaneous cytotoxicity are called natural killer (NK) cells (19), or LGL (large granular lymphocytes) because of the azurophilic granules in their cytoplasm (2). They are effector cells of non-adaptive immunity or natural resistance. In recent years, the anti-microbial activity of NK cells has attracted a great deal of attention (14). Suppression of NK activity along with other components of cellular immunity by an immunosuppressive agent or any other immunosuppressive conditions, might be responsible for fatal infections in the immunocompromised host (e. g. after transplantation) (3). The natural cytotoxicity of lymphocytes on various fungi, including Cryptococcus neoformans (c. neoformans) (15, 20, 21, 22), Coccidioides immitis (C immitis) (16), Paracoccidioides brazilensis (P. braziliensis) (23), Candida albicans (C albicans) have previously been studied in vivo (15) as well as in vitro (16, 24). Natural defence mechanisms are also critical in host resistance against candidiasis both in normal and immunomodulated hosts (17). The aim of this study was to show whether peripheral blood lymphocytes had a natural anti-candidial activity and also to define a new and simple method to assess the NK activity of peripheral blood lymphocytes. Incubation of Candida stellatoidea target cells with lymphocytes resulted in a reduction of the colony forming ability of the yeast. This effect was dependent on the target/effector ratio and incubation period as seen with NK cell cytotoxicity against K562 tumor cell line. After depletion of the adherent cells, the natural anti-candidial activity had not diminished, although the activity was higher when adherent cells were not removed. Healthy male donors were especially chosen because they are thought to be free of infections with pathogenic strains of Candida. Nevertheless, Candida species are also inhabitants of normal flora and the glucan in the yeast wall has strong immunostimulant effects as with other bacterial products (25). In order to show the fraction of lymphocytes that had the natural killer population anti-Leu Ml, anti-Leu llb plus complement were used. While anti-Leu llb (CD16) reacted with 13 % of peripheral lymphocytes and
228 . Z.
GULAy
and T.
IMIR
abolished the ACI, anti-Leu M1 had no significant effect (Table 3). Reduction of ACI and cytotoxicity for both IL-2 activated and non-activated lymphocytes showed that CD 16+ cells are mainly responsible for the lytic effect. Our results correlate with other data in the literature defined as NK antitumoral activity by the classical 4-h-Cr Sl release assay (25). Several techniques for assessing NK activity have been described. BACCARONI et al. (17), described a radio label release micro assay using C. albicans. On the contrary, ZUNINO and HUDDING (24) showed that C. albicans could not be killed by NK cells (showed by CrS1 release assay) and also that NK cell cytotoxicity could be blocked by cold target inhibition when Candida is added on the K562 erythroleucemic cell line. According to our preliminary study, we found that NK activity against Candida yeast cells by using CrS1 release assay was not satisfactory because of the high levels of CrS1 leakage resulted in high spontaneous release and irregular lysing levels even at the same E:T ratio. This observation might be due to the presence of both young and old cells together. Because the Candida aged more than 18 h, caused irregular labelling with CrS1 and lysing. Thus, we used colony formation inhibition assay instead of CrS1 release method to measure NK activity against Candida yeast form. On the other hand, other reports (17, 23) indicate that the hyphal form of Candida could be sensitive to NK cytotoxicity assessed by conventional radiolabel assay. PETKUS et al. (16) reported that C. immitis colony-formation inhibition could be blocked by NK cells. Also, NK cells are effective on C. neoformans colonies both in vitro (20) and in vivo (15). We demonstrated that, NK and LAK cell cytotoxicity on C. albicans could be blocked by simple sugars which was a convincing evidence for the binding process between target and effector. Therefore all yeasts, even Candida species (Table 1) can be susceptible to NK cells lysis, because of their common cell wall sugar moieties. GRIMM et al. (26) first reported that human PBL stimulated with rhIL-2 could develop into non-specific killer lymphocytes termed «lymphokine activated killer (LAK) cells». Several authors investigated the generation of LAK activity against NK resistant target cells and concluded that for the generation of LAK activity PBL should be exposed to rhIL-2 for 3 days, so that Tac receptors appear on PBL. The ability of IL-2 to rapidly induce augmented levels of anti-candidial activity should be independent of the detectable expression of Tac. ORTALDO reported similar results in 1984 for anti-tumoral activity (13). We showed that NK cell activity for ACI was increased with rhlL-2 in dose-dependent manner (Table 4). Pretreatment of PBL with monoclonal antibodies and complement demonstrated that essentially all rhIL-2 responsive cells were confined to the Leu 11 b + subset of lymphocytes and could be blocked by anti-Leu 11 b monoclonal antibody (Table 5). This is also comparable with the results of other authors (4, 27). Our data clearly show that human NK cells have anti-candidial activity in vitro. However, the mechanism of this growth inhibition needs further investigation. The results of this colony forming inhibition assay also
NK and LAK activity on Candida . 229
indicate that, this method is an easy and simple technique that can be used in laboratories with limited financial and technical facilities. Acknowledgements
We gratefully acknowledge the generosity of Prof. OLLI MAKELA (Helsinki University, Finland) for anti-Leu llb and anti-Leu Ml; Prof. ARTUR ULMER (Forschungsinstitut Borstel, Gennany) for anti-IFN-y; Dr. NACI BOR (Hacettepe University, Turkey) for baby rabbit complement. This work was supported by NATO-Science for Stability.Programme
References 1. HERBERMAN, R. B., and H. T. HOLDEN. 1978. Natural cell-mediated immunity. Adv. Cancer Res. 27: 305. 2. TIMoNEN, T., and E. SAKSELA. 1980. Isolation of human natural killer cells by density gradient centrifugation. J. Immunol. Methods. 36: 285. 3. LOTZovA, E., and C. A. SAVARY. 1993. Human natural killer cell development from bone marrow progenitors: analysis of phenotype, cytotoxicity and growth. Nat. Immunol. 12: 209. 4. KARRE, K., M. HANSSON, and R. KIESSLING. 1991. Multiple interactions at the natural killer workshop. Immunol. Today 12: 343. 5. BANKHURST, A. D., and T. IMIR. 1989. The mechanisms involved in the activation of human natural killer cells by staphylococcal enterotoxin B cell. Immunol. 122: 108. 6. SHEN, F., X.-L. Xu, L. H. GRAF, and A. S. F. CHONG. 1995. CD-45 cross-linking stimulates IFN-y production. J. Immunol. 154: 644. 7. D'ORAZIO, J. A., G. W. BURKE, and J. STEIN-STREILEIN. 1995. Staphylococcal enterotoxin B activates purified NK cells to secrete IFN-y but requires T lymphocytes to augment NK cytotoxicity. J. Immunol. 154: 1014. 8. FERRIN], S., A. CAMBIAGGI, S. SFORZINI, S. CANEVARI, D. MEZZANZANICA, M. 1. COLNAGHI, and L. MORETTA. 1993. Use of anti-CD3 and anti-CDl6 bispecific monoclonal antibodies for the targetting of T and NK cells against tumor cells. Cancer Detect. Prevo 17: 295. 9. RILEY, R. S., and E. J. MAHIN. 1989. Clinical application of flow cytometry. ASCP workshop # 9072, Washington DC, p. 35. 10. PERUSSIA, B. 1991. Lymphokine-activated killer cells, natural killer cells and cytokines. Curr. Opin. Immunol. 3: 49. 11. IMIR, T., D. L. GIBBS, W. L. SIBBITT Jr., and A. D. BANKHURST. 1985. Generation of natural killer cells and lymphokine-activated killer cells in human AB serum or fetal bovine serum. Clin. Immunol. and Immunopath. 36: 289. 12. AZZONI, L., 1. ANEGON, B. CALABRETTA, and B. PERUSSIA. 1995. Ligand binding to FcyR induces c-myc dependent apoptosis in IL-2-stimulated NK cells. J. Immunol. 154: 491. 13. ORTALDO, J. R., and J. C. HISERODT. 1989. Mechanism of target cell killing by natural killer cells. Curr. Opin. Immunol. 2: 39. 14. PENARRUBIA, P., F. T. KOSTER, R. O. KELLEY, and A. D. BANKHURST. 1989. Antibacterial activity of human natural killer cells. J. Exp. Med. 169: 99.
230 . Z. GOLAY and T. IMIR 15. LIpSCOMB, M. F., T. ALVARELLOS, G. G. TOEWS, and V. KUMAR. 1987. Role of natural killer cells in resistance to criptococcus neoformans in mice. Am. J. Pathol. 128: 354. 16. PETKUS, A. F., and L. L. BAUM. 1987. Natural killer cell inhibition of young spherules and endospores of Coccidioides Immitis. J. Immunol. 139: 3107. 17. BACCARINI, M., A. VECCHIARELLI, A. CASSONE, and F. BISTONI. 1985. Killing of yeast, germ tube and mycelial forms of C. albicans by murine effectors as measured by a radio label release micro assay. J. Gen. Microbiol. 131: 505. 18. RIPPON, J. W. 1988. Medical Mycology, 3rd ed. W. B. Saunders Company, Philadelphia, 566-575 pp. 19. VERSTEEG, R .. 1992. NK cells and T cells: mirror images? Immunol. Today 13: 244. 20. MILLER, M. F., T. G. MITCHELL, W. J. STORKUS, and J. R. DAWSON. 1990. Human natural killer cells do not inhibit growth of Cryptococcus neoformans in the absence of antibody. Infect. Immun. 58: 639. 21. HmoRE, M. R., N. NABAVI, C. W. REYNOLDS, P. A. HENKART, andJ. W. MURPHY. 1990. Cytoplasmic components of natural killer cells limit the growth of Cryptococcus neoformans. J. Leuc. Biol. 48: 15. 22. LEVITZ, S. M., M. P. DUPONT, and E. H. SMAIL. 1994. Direct activity of human T lymphocytes and natural killer cells against Cryptococcus neoformans. Infect. Immun. 62: 194. 23. JIMENEZ, B. E., and J. W. MURPHY. 1984. In vitro effects of natural killer cells against Paracoccidioides brasiliensis yeast phase. Infect. Immun. 46: 552. 24. ZUNINO, S. J., and D. HUDIG. 1988. Interactions between human natural killer lymphocytes and yeast cells: human NK cells do not kill Candida albicans, although C. albicans block NK lysis of K562 cells. Infect. Immun. 56: 564. 25. BENO, D. W., and H. L. MATTHEWS. 1990. Growth inhibition of Candida albicans by interleukin-2 induced lymph node cells. Cell Immunol. 128: 89. 26. GRIMM, E. A., A. MAZUMDER, H. Z. ZHANG, and S. A. ROSENBERG. 1983. Lymphokine activated killer cell phenomenon. Lysis of natural killer-resistant fresh solid tumor cells by interleukin-2 activated autologous human peripheral blood lymphocytes. J. Exp. Med. 155: 823. 27. DIANZANI, U., D. ZARCONE, V. PISTOIA, C. E. GROSSI, A. PILLERI, M. MASSAIA, and M. FERRARINI. 1989. CD8+/CDllb+ peripheral blood T lymphocytes contain lymphokine"activated killer cell precursors. Eur. J. Immunol. 19: 1037.
Dr. Z. GOLAY, 9 Eyliil University, Faculty of Medicine, Microbiology Department, Inciralti, Izmir, Turkey
©
1996 by Gustav Fischer Verlag, Stuttgart
Department of Microbiology, Gazi University, Faculty of Medicine, Be§evler, Ankara, Turkey
Anti-Candidial Activity of Natural Killer (NK) and Lymphokine Activated Killer (LAK) Lymphocytes In Vitro ZEYNEP GOLAY
and TURGUT IMIR
Received December 8, 1994 . Accepted in revised form October 11, 1995
Abstract The natural cytotoxic effects of peripheral blood lymphocytes (PBL) on Candida stellatoidea and several other Candida species were examined by a colony forming inhibition (CFI) assay. Peripheral blood mononuclear cells (PBMC), were incubated with C. stellatoidea yeast cells. After the incubation period the colony-forming ability of the yeast was significantly reduced. In similar experiments, six different Candida species (c. albicans, C. krusei, C. stellatoidea, C. tropicalis, C. pseudotropicalis, C. guillermondii) were used as target cells. There was no statistically significant difference in the anticandidial activities of PBL against the Candida species used. It was demonstrated that a fraction of lymphocytes, natural killer cells (NK), had the major natural anti-c
Abbreviations: NK = natural killer; LAK = lymphokine activated killer; PBL = peripheral blood lymphocytes; CFI = colony forming inhibition; PBMC = peripheral blood mononuclear cells; mAbs = monoclonal antibodies; JL-2 = saline; FCS = foetal calf serum; IFN-y = interferon-y
NK and LAK activity on Candida . 221
Introduction
NK cells, a component of natural cell-mediated CytotOXICity, are a. heterogeneous population of granular lymphoid cells that lyse certain tumor and virally-infected cells without prior sensitization (1). The NK cell is functionally a subset of lymphocytes (2). NK cells are very heterogeneous with regard to cell surface antigen structure, but with the recent development of monoclonal antibodies (mAbs) the analysis of these antigens could be more comprehensive (3, 4). One of the widely used mAbs for this purpose is anti-CDl6 antibody which binds the low affinity IgG Fe III receptor and reacts with essentially all human NK cells in peripheral blood (5). Anti-CD 16 antibodies recognize 5~0 kDa antigen present on human NK cell membrane and is expressed on approximately 15 % of peripheral blood lymphocytes. Cross-linking of CD16 as well as CD45 (leukocyte common antigen) can stimulate cytokine production (6). It was also reported that CD16 antigen might be expressed on CD 3 positive T lymphocytes in certain individuals (7, 8). It is also expressed on neutrophils, eosinophils, monocytes and macrophage subsets (9). CD15 antigen, the lacto-N-fucopentaeose III molecule, is a branched pentasaccharide with a molecular weight of 185-200 kDa. This carbohydrate moiety can be found in glycolipids and glycoproteins present on cell membrane. It is strongly expressed by neutrophils, eosinophils, monocytes and normal myeloid precursor cells (9). IL-2, a lymphokine produced by T lymphocytes upon stimulation with either mitogens or antigens, is a 15.000 MW glycoprotein in human beings and has a variety of immunoregulatory properties. It is a known activator of NK antitumoral activity (10). Lymphokine-activated killer (LAK) cells, capable of killing malignant target cells in a non-MHC restricted manner, are generated from NK cells upon exposure to IL-2 for 72 h (12). NK cytotoxic activity was first defined for tumor cell lines as they have natural receptors for tumour cell lines such as K562 and Cr51 release assay is used to measure NK cytotoxic activity (13). In recent years, the anti-microbial effect of NK cells has attracted a great deal of attention (14). The in vivo (15) and in vitro (16, 17) effect of NK cells on the various forms of C. albicans has been shown by different methods as indicated in previous reports. In the present study, anti-candidial NK and LAK activity by inhibition of colony formation of various Candida species, is described.
222 . Z. GULAY and T. IMIR
Materials and Methods Fungi
Candida species used throughout this study, were isolated from clinical specimens in our laboratory and identified according to established taxonomic criteria (18). The fungi were grown on a blood agar base for 18 h and the colonies collected while they were on exponential growth term. Yeast cells were resuspended in RPMI 1640 medium (Flow Lab. U.K.) and adjusted to the concentration of 8 x 103 viable organisms per ml. Isolation of peripheral blood lymphocytes (PBL)
Heparinized blood from healthy male donors aged between 20 and 40 was layered on Ficoll-Isopaque separation medium (histopaque 1077, Sigma, St. Louis, Mo, USA) and PBL were isolated by density centrifugation (11). PBL were washed in Phosphate Buffered Saline (PBS) at least three times and resuspended in RPMI 1640 medium supplemented with L-glutamine. The suspension of PBL was poured into Petri dishes sensitized with autologous serum and was incubated for 1 h at 37 °C. After incubation, the non-adherent cells were gently harvested and the same procedure was applied second time. The non-adherent cells were washed in RPMI 1640 and used as the effector cells. The relative percentage of monocytes was less than 5 percent as determined by morphological examination following Giemsa staining. The average cell viability was not less than 98 % as determined by trypan blue dye exclusion test. Cell depletion with mAbs
Anti-Leu lIb (anti-CD 16), anti-Leu Ml (anti-CD 15) monoclonal antibodies (Becton Dickinson, Mountain View, Ca, USA) and baby rabbit complement were used. PBL (5 x 10 6 /ml) was separated into eight groups and each group was incubated with the reagents as follows: 1) anti-Leu 11 b + complement, 2) anti-Leu Ml + complement, 3) anti-Leu I1b+Ml + complement, 4) complement,S) anti-Leu l1b, 6) anti-Leu Ml, 7) only RPMI 1640, 8) IL-2 according to the procedure described earlier (5). Cells were then washed and counted, it was found that 13 % of the total mononuclear cells were CD 16+ and 4 % were CD 15+. The absence of any residual cells reacting with the relevant antibodies after complement-mediated lysis was verified by immunofluorescence. The PBLs were counted, their number corrected and used as effectors in the assay. Effect of IFN-y on cytotoxicity
In order to investigate whether the cytotoxic activity on colony inhibition was a reason for IFN-y, we used anti-IFN-y monoclonal antibody (Boehringer Mannheim, Germany) to neutralize if any IFN-y which is secreted mostly by NK cells. After addition of 10 f!g/ml anti-IFN-y to each well, inhibitory effect of NK cells was measured using the anticandidial cytotoxicity assay. Activation with recombinant interleukin-2 (IL-2)
rhL-2 (Sigma, St. Louis, MO, USA) was used in various concentrations from 3 to 100 U I
ml for 106 lymphocytes. Monocyte depleted PBL were incubated with rhL-2 for 72 hand used as lymphokine activated killer (LAK) cells (11). The viability o f PBLs after incubation was always more than 90 % determined by trypan blue dye exclusion.
NK and LAK activity on Candida . 223 Colony forming inhibition assay (CFI) C. stellatoidea yeast cells, used as target cells (unless otherwise stated), were incubated with effector cells in 96-well U-bottomed microtitre plates at different target:effector ratios (0.1 ml of a 8 x 103 yeast cells/ ml Candida suspension with 0.1 ml of effector cell suspensions) for 2 h at 37 °C in a 5 % CO 2 incubator (16). Target cells, incubated in RPMI 1640 medium alone were used as controls. After incubation, aliquots of 25 [.ll from each tube were plated on Petri dishes containing brain heart infusion agar. Cultures were then incubated for 48 h at 37°C, colonies were counted (100 colonies per plate expected on control plates) and the percentage reduction in colony numbers was determined as the Anti-Candidial Index (ACI) according to the following formula:
ACI= ( 1-
experimental colony number) x 100 control colony number
Each experiment was performed in quadruplicate. The results were represented as the mean ± SEM values of ACI obtained from separate groups. Chromium release assay for NK and LAK cell activity
K562 (for the NK assay) and Daudi (for the LAK assay) cells were labelled by incubating 5 x 106 target cells with 0.1 ml Na/1Cr04 (specific activity 1.0 mCi/ ml, Amersham, U.K.) for 1 h at 37"C in a humidified air plus 5 % CO 2, washed three times and then 10 4 target cells in 0.1 ml was used in each micro titre well. Effector cells, at a density of 25 times the target cell number (E:T 25:1) in 0.1 ml volume, were added to labelled cells in the wells. Spontaneous release was estimated by incubating target cells with 0.1 ml medium only. Maximum release was determined by incubating labelled target cells with 0.1 ml of 2 % sodium dodecyl sulphate. After a 4-h incubation period, the microtitre plates were centrifuged at 150 x g for 5 min and 0.1 ml of supernatants from each well was counted in a gamma counter (LKB, 1275 minigamma). Cytotoxicity was calculated from the average counts per minute (cpm) released into the supernatants of triplicate samples using the following formula (5): % specific cytotoxicity =
test cpm - spontaneous cpm x 100 max. cpm - spontaneous cpm
Statistical ana'lysis
The results are expressed as the mean ± SEM. The significant difference between two means was determined by the Student's t-test.
Results Anti-candidial effect of lymphocytes
When freshly isolated and monocyte-depleted human PBL from healthy male donors was incubated for 1, 2 and 4 h with C. stellatoidea, the ability of Candida to form colonies was inhibited. Representative results obtained from four donors are presented in Figure 1. The spontaneous cytotoxicity of PBL against C. stellatoidea from different donors varied somewhat but
224 . Z.
GOLAY
and T.
IMIR
60
1:5
1:10 1:15
1:5
1:10 1:15
1:5
1:10 1:15 T:E
1h
2h
4h
Incubation Time Figure 1. PBL's anti-candidial effects on C. stellatoidea. The standard error of all determinants was less than 7 %. n = 4.
all donors demonstrated anti-candidial activity. The level of inhibition of colony formation (anti-candidial index), was dependent on effector cell number and incubation period. Longer incubation of effector/target mixture resulted in higher inhibition. Target: effector cell ratios of 1:5, 1: 10 and 1:15 were used for ACI. After 1 h incubation period, the mean ± SEM values of % inhibition in colony-forming were 9.5 ± 4.8, 15.5 ± 5.1 and 26.2 ± 6.7 for T:E ratios of 1:5, 1:10 and 1:15, respectively. For 2 h incubation period: 37.2 ± 3.7, 45.9 ± 5.8 and 53.6 ± 5.8 and for 4 h incubation period: 48,8 ± 2.2, 56,0 ± 4,5 and 62.5 ± 6.0 percent inhibition were achieved at the same T:E ratios.
Table 1. The mean values of PBL's ACI for different species. 2 h ACI activity was performed at the T:E ratio of 1:10. Data were expressed as the mean ± SEM, n=4.
C. C. C. C. C. C.
albicans stellatoidea tropicalis pseudotropicalis krusei guillermondii
28.3 ± 3.4"29.9±2.7"" 28.6 ± 1.6':29.4 ± 2.5" 32.2 ± 2.2" 32.1 ± 1.0':-
NK and LAK activity on Candida . 225 Table 2. The effect of anti IFN-y on ACI. The experiment is representative of three performed, 'T:E= 1:10, b mean ± SEM. ACI a 38.4 ± 3.6 b 36.1 ± 4.3
Control (C) C + anti-IFN-y
Different Candida species and ACI activity
ACI of effector cells against 6 different Candida species (c. stellato idea, C. albicans, C. tropicalis, C. pseudotropicalis, C. guillermondii, C. krusei) was compared in 7 different experiments using the same effector cells obtained from one donor. Although some strains are established pathogens (for example C. albicans), it is important to state that all the species used were isolated from clinical specimens. At the target:effector ratio of 1: 10 no statistically significant difference in ACI was noted among different Candida species (Table 1). The effect of anti-IFN-y, anti-Leu 11 b, anti-Leu M 1 and complement pretreatment of PBL an ACI
In this experiment anti-IFN-y was used as a cytokine inhibitor to show that the cytotoxic mechanism is not due to IFN -y secretion. As is known that it has an activating effect on NK cells to lyse target cells, anti-IFN-y was present during 4 h of incubation period. The results demonstrated that there was no measurable inhibition of NK activity (Table 2). Table 3. The effect of monoclonal antibodies and complement on ACI (ACI values are expressed as mean ± SEM, the numbers in parentheses correspond to the % inhibition in ACI compared with control PBL in RPMI 1640 only). The cell viabil ity was> 95 % in all cultures and this did not change with or without exposure to mAbs and/ or complement. Specific cell populations indicated were removed by complement-mediated lysis. C. stellatoidea was used as target cells. Data were represented as the mean ± SEM of three different experiments. ACI
1:5 Control (C) C + Complement (C') anti-Leu 11 b anti-Leu Ml + C' anti-Leu 11 b + C' anti-Leu M1 + anti-11 b + C'
26.7 ± 3.2 27.2 ± 2.5 25.7 ± 6.0 20.6 ± 2.0 3.8 ± 1.4 1.5 ± 2.1
1:15
1:10
(0) (4) (23) (86) (95)
43.5 ± 6.4 45.5 ± 2.8 42.3 ± 4.6 36.5 ± 6.0 9.0 ± 5.4 1.6 ± 3.8
(0) (0) (16) (79) (96)
49.0 ± 6.9 51.5 ± 5.2 (0) 49.5 ± 10.6 (0) 41.5 ± 7.2 (15) 10.7 ± 5.0 (78) 5.0 ± 6.9 (90)
226 . Z.
GULAy
and T. IMIR
Table 4. Enhancement of anti-candidial activity by rhIL-2." Activation relative to untreated control, a PBL incubated with IL-2 for 72 h, b mean values ± SEM, CP < 0.05, the experiment shown is representative of three performed. ACI (T:E) IL-2 Concentration (U/ml/l06 PBL)
o (Control)a 3 10 30 100
1:5 (%)*
1:10 (%)
1:15 (%)
24.1 ± 2.9 b 27.8 ± 6.6 (15.4) 40.2 ± 5.2 (66.8) 43.0 ± 4.8 (78.4)C 42.0 ± 14.8 (74.3)C
31.7 ± 4.4 36.7 ± 2.3 (15.7) 47.6 ± 6.1 (49.9) 52.7 ± 5.8 (66.0)C 54.0 ± 11.2 (70.0t
39.5 ± 2.2 45.8 ± 6.3 (15.9) 55.1 ± 4.0 (39.6) 60.9 ± 7.6 (54.2t 64.4 ± 13.9 (58.4)"
To show which subpopulation of lymphocytes has natural anti-candidial activity, anti-Leu llb, anti-Leu Ml monoclonal antibodies plus complement were used. In the anti-Leu lIb + Leu Ml + C'group, anti-candidial activity was reduced by 96 % (p < 0.01) (Table 3). ACI was not changed either with complement or anti-Leu 11 b group, but strong inhibition was observed when both components were used together. Anti-Leu Ml antibody plus complement diminished ACI insignificantly. Therefore, it was concluded that most natural anti-candidial activity was performed in the NK cell population. Enhancement of anti-candidial activity by rhlL-2
Adherent, cell-depleted PBL from 4 healthy donors was incubated for 72 h with 0 (control), 3,10,30 and 100U/ml rhIL-2, and then assayed for anticandidial activity. As seen in Table 4, effect of rhIL-2 was dose-dependent, the magnitude of response increased proportionally with the higher doses of rhIL-2. Three days' incubation of PBL with rhIL-2, strongly stimulated Table 5. Inhibition of NK activity on ACI and cytotoxicity by mAb + c. aT:E = 1:5, bT:E= 1:25, cPBL incubated with and without 100 Vlml IL-2 for 72 h, d mean ± SEM, e PBL treated with mAb + C, as described in Materials and Methods. The representative of 4 experiments is shown.
Target cells
ACP
% Cytotoxicityb
C. stellatoidea
K562
HIL2 Control PBL PBL + anti-Leu 11 b + C e
(+) IL2 C
HIL2
Daudi (+) IL2
HIL2
(+) IL2
15.2 ± 3.5d 27.6 ± 4.9 21.8 ± 5.2 37.3 ± 6.4 12.3 ± 4.4 34.2 ± 5.8 5.8 ± 4.1
6.5 ± 3.2
8.4 ± 3.7
9.8 ± 5.7
7.4 ± 3.3 11.7 ± 5.6
NK and LAK activity on Candida . 227
the natural anti-candidial activity. Adherent cell-depleted effector cells were pre-treated with anti-Leu llb, anti-Leu Ml and complement as described in Materials and Methods prior to stimulation with rhIL-2. As seen in Table 5, the NK activity on C stellatoidea was significantly reduced after treatment with mAbs plus complement and the enhancing activity of rhIL-2 was nearly abolished. Effector cells were incubated with 100 V/ml rhIL-2 for 72 h before the ACI assay.
Discussion A subpopulation of human peripheral blood lymphocytes from healthy donors are able to lyse a variety of target cells, including tumor and virusinfected cells, without prior sensitisation. The effector cells responsible for this spontaneous cytotoxicity are called natural killer (NK) cells (19), or LGL (large granular lymphocytes) because of the azurophilic granules in their cytoplasm (2). They are effector cells of non-adaptive immunity or natural resistance. In recent years, the anti-microbial activity of NK cells has attracted a great deal of attention (14). Suppression of NK activity along with other components of cellular immunity by an immunosuppressive agent or any other immunosuppressive conditions, might be responsible for fatal infections in the immunocompromised host (e. g. after transplantation) (3). The natural cytotoxicity of lymphocytes on various fungi, including Cryptococcus neoformans (c. neoformans) (15, 20, 21, 22), Coccidioides immitis (C immitis) (16), Paracoccidioides brazilensis (P. braziliensis) (23), Candida albicans (C albicans) have previously been studied in vivo (15) as well as in vitro (16, 24). Natural defence mechanisms are also critical in host resistance against candidiasis both in normal and immunomodulated hosts (17). The aim of this study was to show whether peripheral blood lymphocytes had a natural anti-candidial activity and also to define a new and simple method to assess the NK activity of peripheral blood lymphocytes. Incubation of Candida stellatoidea target cells with lymphocytes resulted in a reduction of the colony forming ability of the yeast. This effect was dependent on the target/effector ratio and incubation period as seen with NK cell cytotoxicity against K562 tumor cell line. After depletion of the adherent cells, the natural anti-candidial activity had not diminished, although the activity was higher when adherent cells were not removed. Healthy male donors were especially chosen because they are thought to be free of infections with pathogenic strains of Candida. Nevertheless, Candida species are also inhabitants of normal flora and the glucan in the yeast wall has strong immunostimulant effects as with other bacterial products (25). In order to show the fraction of lymphocytes that had the natural killer population anti-Leu Ml, anti-Leu llb plus complement were used. While anti-Leu llb (CD16) reacted with 13 % of peripheral lymphocytes and
228 . Z.
GULAy
and T.
IMIR
abolished the ACI, anti-Leu M1 had no significant effect (Table 3). Reduction of ACI and cytotoxicity for both IL-2 activated and non-activated lymphocytes showed that CD 16+ cells are mainly responsible for the lytic effect. Our results correlate with other data in the literature defined as NK antitumoral activity by the classical 4-h-Cr Sl release assay (25). Several techniques for assessing NK activity have been described. BACCARONI et al. (17), described a radio label release micro assay using C. albicans. On the contrary, ZUNINO and HUDDING (24) showed that C. albicans could not be killed by NK cells (showed by CrS1 release assay) and also that NK cell cytotoxicity could be blocked by cold target inhibition when Candida is added on the K562 erythroleucemic cell line. According to our preliminary study, we found that NK activity against Candida yeast cells by using CrS1 release assay was not satisfactory because of the high levels of CrS1 leakage resulted in high spontaneous release and irregular lysing levels even at the same E:T ratio. This observation might be due to the presence of both young and old cells together. Because the Candida aged more than 18 h, caused irregular labelling with CrS1 and lysing. Thus, we used colony formation inhibition assay instead of CrS1 release method to measure NK activity against Candida yeast form. On the other hand, other reports (17, 23) indicate that the hyphal form of Candida could be sensitive to NK cytotoxicity assessed by conventional radiolabel assay. PETKUS et al. (16) reported that C. immitis colony-formation inhibition could be blocked by NK cells. Also, NK cells are effective on C. neoformans colonies both in vitro (20) and in vivo (15). We demonstrated that, NK and LAK cell cytotoxicity on C. albicans could be blocked by simple sugars which was a convincing evidence for the binding process between target and effector. Therefore all yeasts, even Candida species (Table 1) can be susceptible to NK cells lysis, because of their common cell wall sugar moieties. GRIMM et al. (26) first reported that human PBL stimulated with rhIL-2 could develop into non-specific killer lymphocytes termed «lymphokine activated killer (LAK) cells». Several authors investigated the generation of LAK activity against NK resistant target cells and concluded that for the generation of LAK activity PBL should be exposed to rhIL-2 for 3 days, so that Tac receptors appear on PBL. The ability of IL-2 to rapidly induce augmented levels of anti-candidial activity should be independent of the detectable expression of Tac. ORTALDO reported similar results in 1984 for anti-tumoral activity (13). We showed that NK cell activity for ACI was increased with rhlL-2 in dose-dependent manner (Table 4). Pretreatment of PBL with monoclonal antibodies and complement demonstrated that essentially all rhIL-2 responsive cells were confined to the Leu 11 b + subset of lymphocytes and could be blocked by anti-Leu 11 b monoclonal antibody (Table 5). This is also comparable with the results of other authors (4, 27). Our data clearly show that human NK cells have anti-candidial activity in vitro. However, the mechanism of this growth inhibition needs further investigation. The results of this colony forming inhibition assay also
NK and LAK activity on Candida . 229
indicate that, this method is an easy and simple technique that can be used in laboratories with limited financial and technical facilities. Acknowledgements
We gratefully acknowledge the generosity of Prof. OLLI MAKELA (Helsinki University, Finland) for anti-Leu llb and anti-Leu Ml; Prof. ARTUR ULMER (Forschungsinstitut Borstel, Gennany) for anti-IFN-y; Dr. NACI BOR (Hacettepe University, Turkey) for baby rabbit complement. This work was supported by NATO-Science for Stability.Programme
References 1. HERBERMAN, R. B., and H. T. HOLDEN. 1978. Natural cell-mediated immunity. Adv. Cancer Res. 27: 305. 2. TIMoNEN, T., and E. SAKSELA. 1980. Isolation of human natural killer cells by density gradient centrifugation. J. Immunol. Methods. 36: 285. 3. LOTZovA, E., and C. A. SAVARY. 1993. Human natural killer cell development from bone marrow progenitors: analysis of phenotype, cytotoxicity and growth. Nat. Immunol. 12: 209. 4. KARRE, K., M. HANSSON, and R. KIESSLING. 1991. Multiple interactions at the natural killer workshop. Immunol. Today 12: 343. 5. BANKHURST, A. D., and T. IMIR. 1989. The mechanisms involved in the activation of human natural killer cells by staphylococcal enterotoxin B cell. Immunol. 122: 108. 6. SHEN, F., X.-L. Xu, L. H. GRAF, and A. S. F. CHONG. 1995. CD-45 cross-linking stimulates IFN-y production. J. Immunol. 154: 644. 7. D'ORAZIO, J. A., G. W. BURKE, and J. STEIN-STREILEIN. 1995. Staphylococcal enterotoxin B activates purified NK cells to secrete IFN-y but requires T lymphocytes to augment NK cytotoxicity. J. Immunol. 154: 1014. 8. FERRIN], S., A. CAMBIAGGI, S. SFORZINI, S. CANEVARI, D. MEZZANZANICA, M. 1. COLNAGHI, and L. MORETTA. 1993. Use of anti-CD3 and anti-CDl6 bispecific monoclonal antibodies for the targetting of T and NK cells against tumor cells. Cancer Detect. Prevo 17: 295. 9. RILEY, R. S., and E. J. MAHIN. 1989. Clinical application of flow cytometry. ASCP workshop # 9072, Washington DC, p. 35. 10. PERUSSIA, B. 1991. Lymphokine-activated killer cells, natural killer cells and cytokines. Curr. Opin. Immunol. 3: 49. 11. IMIR, T., D. L. GIBBS, W. L. SIBBITT Jr., and A. D. BANKHURST. 1985. Generation of natural killer cells and lymphokine-activated killer cells in human AB serum or fetal bovine serum. Clin. Immunol. and Immunopath. 36: 289. 12. AZZONI, L., 1. ANEGON, B. CALABRETTA, and B. PERUSSIA. 1995. Ligand binding to FcyR induces c-myc dependent apoptosis in IL-2-stimulated NK cells. J. Immunol. 154: 491. 13. ORTALDO, J. R., and J. C. HISERODT. 1989. Mechanism of target cell killing by natural killer cells. Curr. Opin. Immunol. 2: 39. 14. PENARRUBIA, P., F. T. KOSTER, R. O. KELLEY, and A. D. BANKHURST. 1989. Antibacterial activity of human natural killer cells. J. Exp. Med. 169: 99.
230 . Z. GOLAY and T. IMIR 15. LIpSCOMB, M. F., T. ALVARELLOS, G. G. TOEWS, and V. KUMAR. 1987. Role of natural killer cells in resistance to criptococcus neoformans in mice. Am. J. Pathol. 128: 354. 16. PETKUS, A. F., and L. L. BAUM. 1987. Natural killer cell inhibition of young spherules and endospores of Coccidioides Immitis. J. Immunol. 139: 3107. 17. BACCARINI, M., A. VECCHIARELLI, A. CASSONE, and F. BISTONI. 1985. Killing of yeast, germ tube and mycelial forms of C. albicans by murine effectors as measured by a radio label release micro assay. J. Gen. Microbiol. 131: 505. 18. RIPPON, J. W. 1988. Medical Mycology, 3rd ed. W. B. Saunders Company, Philadelphia, 566-575 pp. 19. VERSTEEG, R .. 1992. NK cells and T cells: mirror images? Immunol. Today 13: 244. 20. MILLER, M. F., T. G. MITCHELL, W. J. STORKUS, and J. R. DAWSON. 1990. Human natural killer cells do not inhibit growth of Cryptococcus neoformans in the absence of antibody. Infect. Immun. 58: 639. 21. HmoRE, M. R., N. NABAVI, C. W. REYNOLDS, P. A. HENKART, andJ. W. MURPHY. 1990. Cytoplasmic components of natural killer cells limit the growth of Cryptococcus neoformans. J. Leuc. Biol. 48: 15. 22. LEVITZ, S. M., M. P. DUPONT, and E. H. SMAIL. 1994. Direct activity of human T lymphocytes and natural killer cells against Cryptococcus neoformans. Infect. Immun. 62: 194. 23. JIMENEZ, B. E., and J. W. MURPHY. 1984. In vitro effects of natural killer cells against Paracoccidioides brasiliensis yeast phase. Infect. Immun. 46: 552. 24. ZUNINO, S. J., and D. HUDIG. 1988. Interactions between human natural killer lymphocytes and yeast cells: human NK cells do not kill Candida albicans, although C. albicans block NK lysis of K562 cells. Infect. Immun. 56: 564. 25. BENO, D. W., and H. L. MATTHEWS. 1990. Growth inhibition of Candida albicans by interleukin-2 induced lymph node cells. Cell Immunol. 128: 89. 26. GRIMM, E. A., A. MAZUMDER, H. Z. ZHANG, and S. A. ROSENBERG. 1983. Lymphokine activated killer cell phenomenon. Lysis of natural killer-resistant fresh solid tumor cells by interleukin-2 activated autologous human peripheral blood lymphocytes. J. Exp. Med. 155: 823. 27. DIANZANI, U., D. ZARCONE, V. PISTOIA, C. E. GROSSI, A. PILLERI, M. MASSAIA, and M. FERRARINI. 1989. CD8+/CDllb+ peripheral blood T lymphocytes contain lymphokine"activated killer cell precursors. Eur. J. Immunol. 19: 1037.
Dr. Z. GOLAY, 9 Eyliil University, Faculty of Medicine, Microbiology Department, Inciralti, Izmir, Turkey