Document gaOLOnMmXwaLQ722JRJQnb4ee

LrmtrmimXr--r*VC frtwlW la C<*4 Brit*. UVU7, IM4. 0.00 O IH< f a i t iwa FYcu U 4,' NATURAL KILLER (NK)-CELL ACTIVITY AND ANTIBODY-DEPENDENT CELLULAR CYTOTOXICITY (ADCQ IN PRIMARY PRELEUKEMIC SYNDROME* CiTTE Kerndrup, Karin Meyer, J^ rcen Ellegaard and Peter Hokland University Department o f Medicine and Haematology, Aarhus Amtssygehus, Aarhus, Denmark (Received4 May 1983. Revision accepted 8 August 1983) Abstract--In 13 patients with a multi-parameter based diagnosis o f primary acquired preleukemic syndrome (PPS), natural kilter (NK) cdl activity and antibody-dependent cellular cytotoxicity (A DCQ were investigated on peripheral blood mononuclear cell (MNC) fractions, lo all but two patients a defective NIC activity was found. Lymphocyte-monocyte mixture experiments demonstrated that (his was not due to suppressor monocytes. Furthermore, NK activity was defec tive when both mydoid and non-mydoid target cdl lines were used. Addition of human leukocyte interferon to the NK cultures augmented the cytotoxicity, which exhibited the same kinetics as that o f normal controls, but NK activity Icvds remained subnormal. These data strongly indicate that the decreased NK activity seen in the patients is due to a decreased number of circulating NK cells. In contrast ADCC was within the normal range both when MNC suspensions as well as when purified peripheral blood lymphocytes were used as effector cells thus ruling out subnormal lymphocyte ADCC masked by the presence of monocyte ADCC. These results demonstrate that PPS patients have a selective NK defect with an intact lymphocyte ADCC function. Whether this defect will prove to be valuable in the assessment of a malignant transformation in a given patient will await further longitudinal NK studies and dinical follow-up of the patients. Key words: Prdeufcemic syndrome, NK activity, ADCC, a-Interfcron. INTRODUCTION PRIMARY acquired prclcukcmic syndrome (PPS) is, in spite of intensive investigation, still an iU-defined entity when diagnosis is based on dinical and morphological criteria alone [2, 5, 16, 20, 26, 271- Linman [16] defined PPS as an unexplained mono-, bi- or pancytopenia, often accompanied by a hypercdlular bone marrow with megaloblastic morphological changes. With respect to bone marrow morphology PPS can be defined as a condition with less than 5% bone marrow myeloblasts and it is thus distinguished from smouldering leukemia or refractory anaemia with an excess of blasts, in which the per centage of myeloblasts by definition exceeds 5% but is less than 209 [2]. Since natural killer (NK) cells might play a role in the host defence against early neoplasia [18] and since some patients with PPS later develop acute myelogenous leukemia (AML) [15, 27], the purpose of this study was to elucidate whether patients with a diagnosis of PPS, as defined above, had significant alterations in NK activity. In normal human subjects and in experimental animal systems, the NK activity is, both in vivo and in vitro, enhanced by interferon or .interferon inducers [6, 11, 13, 30]. In consequence of this, we also investigated the effect of human leukocyte interferon (a-IFN) on NK activity in vitro. The conflicting results concerning antibody-dependent cellular cytotoxicity (ADCQ in the defence against neoplasia [4, 22] likewise led us to investigate possible changes in ADCC in patients with PPS. Supported by the Danish Research Council Grant No. 12-3132. Abbreviations; ADCC. antibody-dependent cellular cytotoxicity; a-IFN, human leukocyte inlerfcron; A M L, acute myelogenous leukemia; e ft. effector to target; MNC, mononuclear cell; NK, natural killer; PPS, primary preleukonk syndrome. Correspondence to; Dr Gitle Kerndrup. Mcdicinsk afdcting 11. Aarhus Amtuygchus, DK-SOOO Aarhui C, Denmark. 239 240 Citte Kckndkup et al. MATERIALS AND METHODS Patients The material comprise* I] patients, five women and eight men, aged 36-84 yean. The patients were admitted to the University Department o f Medicine and Haematology. Aarhtu Amtuygchus, during the period March 1979-October 1981. None had received cytostatics or radiotherapy prior to admission. Diagnosis was based oa several abnormalities regarding clinical, chemical and morphological findings in each pa dent. The diagnostic criteria were unexplained mono*, bi- or pancytopenia of the blood with simultaneous presence o f a hypcrcdlular bone marrow, (n addition, a number o f other features were present in most patients at the dme of diagnosis (Table I). Alt patients had karyotypic abnormalities when examined by banding technique. All patients bad less than 3* myd oblasts in thdr bone marrow and none had rayd oblasts in peripheral blood at time o r diagnosis as wdl as at time o f Investigation, At the end o f the observation period four patients had died (one patient of overt AML, one of septicemia, one of ischaemic colitis and one o f pneumonia). Natural kilter cell assay M ononudcar effector cdIs were isolated on Isopaque-Flcoll gradients (Lymphoprep*, Nygaard, Oslo) [3.12]. As target cells the NKfcnsitivc crythroleukenua cd l line K562 was used [12], Culture conditions for the J,Cr rdeasc assay were as described in detail previously (12] and test systems were set up with and without the addition of 500 1U of partially purified a-IFN (kindly provided by Dr Kurt Berg, Institute o f Microbiology, University of Aarhus). Spedfic lysis o f target cells was calculated from the formula: Counts/min Exp - Counts/min S R * , Counts/min MR - Counts/min SR T a b l e I. In i t i a l b l o o o a n o b o n e m a x r o w fin d in g s in 13 p a t ie n t s WITH A DIAGNOSIS O f PRIMAXY ACQUIREO PtELEUKEWIC SYNDROME Variable Result Sex Mcan agc Mcan obscrvation lime (months) Responsc to treatment 6F/7M 66 (range: 36-84) 32 (range: 3-144)f l/llt Blood findings Anaemia (F < 7 .3 m m o t/l; M < 8 ,2 mdvol/l) 13/13 Macrocytoss (M CV >I02 fi) 6/13 Granulocytopenia ( < 2 X IO1/! ) 9/13 Thrombocytopenia (< 3 0 x 10*/l) 5/13 Pancytopenia 3/13 S-cobaiamna I (normal: 200-800 pm ol/l) 2/13 Granulocyte hypoicjmentation Karyotypic abnormalities ------ - 8 / 1 3 -------- ----ll/ll Bone marrow findings Ovcrali cdlularity (+ + + / + + /+ . / N / --/ ----- / -------- ) Erythropoietic cdlularity (4- + / + / N / --/ ----- ) Granulopoietic cdlularity (i- * / + / N / --/ ----- ) Bone marrow iron content 1 Sidcroblasis 1 Mydoblasts ( < 5 * ) 7 /4 /0 /I/1 /0 /0 3/7/2/1/0 3/8/1/I/O 12/13 10/121 13/13 Included are the dinical data at time of diagnosis and the treatment. ' F - female. M - male. fFour patients have died at end o f observation lime. Response to treatment (folinic ad d , pyridoaine. anasteronc and/or prednisone) measured in terms of diminished requirement for blood transfusions. tin one case smear preparations of bone marrow were not available for revision. where cour cultured, ct only, and t results were as routine i. each espcri In select* leukemia) I: Depletion a The mon* adherence r staining for spedfic for percentage purified me ADCC Except fc assays. As b without the tion of spet compared t* In Tour p; c /t ratios ai Periphc tested for tests. As \ that of 20 F*g . NK edit and lymphocyte ADCC tn preieukemia 241 where counu/m in Exp U the radioactivity of the supernatant when effector ed it and target e d it were co cultured, countt/m in SR the spomancous release of radioactivity from cultures containing labelled Urget ed it only, and countt/m in MR the maximal rdcase determined after detergent lytii of labelled target ed it. The result! were compared to a panel o f normal controls (n * 20). According to earlier obtervations 16 h was chosen a* routine incubation period [I2|. Different effector to urget (e/t) ratios (12.3:1, 23:1, 50:1, 100:1) were used in each experiment. In sdected experiments a pand of NK-tensitive Urget ed it induding the 1301 (acute T-lymphoblastic leukemia) line, and the MOLT-4 (acute Trlymphoblastk leukemia) line, was employed. Dcfyictio* */*d purification o f monocytes The mononuclear ceils Isolated from the Mood were separated into lymphocytes and monocytes by the plastic adherence method as previously described in detail (I5J. Purity of the isolated suspensions was estimated after turning for non-specific esterase etuyme activity and by reactivity with the monodonal antibody Mo2, which is specific for monocytes (29). Monocyte contamination o f the lymphocyte suspension never exceeded 2*?, and the percentage o f monocytes in the adherent cdt fraction in all insunccs exceeded 6 0 S . In selected experiments, purified monocytes were added in graded numbers to NK cultures. ADCC Except for the urget cells and the o-lFN addition, the ADCC cultures were set up simitar to the NK activity assays. As u rget cells the mastocytoma cdl line P815 was used (e/t ratio 30:1). Assays included cultures with and without the addition of rabbit anti-PSI5 anti-serum (final dilution 1:2300). Incubation, harvesting and calcula tion o f specific chromium release was done as described for the NK activity assays. The results were likewise compared to a panel of normal controls (n - 20). In four patients ADCC assays were carried out on purified lymphocyte and monocyte cell fractions at various e/t ratios and the results compared to results obtain! with the unseparated mononudear cell fractions. RESULTS Peripheral blood mononuclear cells isolated from 13 patients with prclcufcemia were tested for spontaneous as well as antibody-dependent cytotoxicity in the NK and ADCC tests. As will be seen from Fig. 1, in which the NK activity of the patients is compared to that of 20 normal donors, only two of the 13 patients had a normal level of NK activity. % SPtCIflC * k r IElAS Fig. I. NK activity of 13 prelcukcmia patients both with and without the addition of 300 units of a-IFN . The dotted area indicates the normal range S.D. 242 Gim KetNCxturetet. After addition of 500 units of a-IFN to the cultures, an increase in the NK activity com parable to that of the normal donors was observed. However, only one patient remained within the normal range, indicating that the unstimulatcd, but not necessarily the interferon activated NIC cells, possessed severely depressed cytotoxic activity. In order to evaluate whether the depressed NK activity was due to a decreased level of NK activity by a normal number of circulating NK cells, or alternatively to a decreased number of NK cells in the peripheral blood, different effector-to-target ratios were used in all assays. From Fig. 2, which shows the results from a representative experiment, it is evident that the kinetics of K562 target cell lysis are similar in the patient and the normal donor. This may indicate that the defective NK activity is due to a reduced number of NK cells in the blood and not to an intrinsic defect in the NK cells. Fto. 2. Effcctor/targct ratio curves of a representative normal donor ( ) and a prdeukem ia patient (o) in NK assay against K56Z edit (dosed symbols, after the addition of 500 IU a-IFN). To elucidate further the NK defect, NK activity assays were carried out with two T-ALL cell lines as target cells,, simultaneously with the routinely used K562 target cell Une. Decreased NK activity was also demonstrated against these target cells (Fig. 3). Since the mononuclear cell fraction harvested from the gradient interface contained from 5 to 25% monocytes (as detected by the monoclonal antibody Mo2), we found it im portant to evaluate the influence of monocytes upon NK activity and ADCC. In the NK assay, where monocytes in high pcrcenlgs"couId be imagined to inhibit cytotoxicity,lymphocytes and monocytes were separated and graded numbers of purified monocytes were added lo the fixed numbers oflymphocytes and target cells in NK cultures. As will be seen from Fig. 4, addition of up to 40% monocytes did not significantly influence the-NK activity of either patient or normal donor lymphocytes (monocytes alone gave no lysis). Thus, it seems unlikely that the decreased NK activity of the preleukcmia patient is due to the presence of peripheral blood suppressor monocytes. in contrast to the NK results Fig. 5 shows that nine patients had normal ADCC and only three had decreased cytotoxic activity against sensitized P815 mastocytoma cells. In one patient a significantly increased cytotoxic activity was found. In the ADCC assay using IgG-scnsittzcd P815 cells, which arc also susceptible to lysis by monocytes, we tested the effect of monocyte depletion of the effector cell suspensions. In Tabic 2 data from four such experiments are given and it is apparent that non-adherent lymphocytes from the patients retained their cytotoxic activity in the ADCC assay, indicating that the ADCC results (Fig. 5) are not due to cytotoxic monocytes, but can be attributed to the patient lymphocytes only. Fkj. 3. NK activity of three prdeukemia patients (right columns) and three normal donors ( r S.O.) against three different target cdl lines without a-IFN and with (total columns) the addition of 500 a-IFN units/ml. L CL ) a id eo O. b . Q. b . Fie. 4. NK actWhy of peripheral blood lymphocytes from prdeukemia patient (left panel) and a normal donor (rifht panel) after addition of traded numbers of monocytes (monocytes alone did not result In lysis). Cytotoxicity without a-IFN and with (total columns) the addition of 500 a-IFN units/m l. x S.D. is shown for simultaneous assays En each experiment. 244 Gifte KcRNORUr tat. % PfCIMC-'cr Mtllit 1 1 so- NORMALS (n*201 -SD i oH ioH Fig. 5. Antibody-dependent cellular cytotoxicity (ADCC) io 13 prdeukemia patients compared to that o f normal controls (dotted area). The figure depicts the specific chromium release with and without the addition of an anti-PS 15 antiserum (final dilution 1:2500). T able 2. A D C C of unseparated and separated cell fractious in four patients wit primary preleucemic * SYNDROME E /t m o 12.5:1 25:1 50:1 a>ft Patient No. 4 Patient No. II Patient No. 42 Patient No. I U nsep.t N on-adh.t Uruep. Non-adh. Unxep. Non-adh. Unxep. Noo-tdh, 2.21 2.6 0.8 0.5 0.3 2.5 --0.5 1.2 13.5 14.2 45.1 35.3 13.0 23.3 26.3 2 7 J _ O.S 2.8 0.6 2.8 --2.9 9.8 0.0 2.6 + 21.0 22.5 57.9 54.5 32.9 36.3 42.8 40.3 -- 4.9 5.3 0.7 1.6 --3.3 21.6 n.t. 5.2 + 36.0 32.6 5S.0 60.7 40.2 43.1 n.t. 49.9 ' Addition o f anti-PS 15, final dilution 1:2500. flsopaque-Ficotl separated mononudear cells. PIa*tic non-adherent odls. 1Median value of triplicate determination, n.l.: Not texted. DISCUSSION Cell-mediated immune reactions have been thought to play an important role in the rejection of experimental tumors or grafts. The effector cell mediating these cytotoxic reactions was originally thought to be the cytotoxic T lymphocyte (CTL). Whereas serious doubts ha* cells arc n surveillaw -review, set with the lymphom; found (9, individual of which 14]. An it malignan growth. I capacity investigai against c activity in 11 significar activity c defect of was simi monocyt amount non-myc defect ij with a * showed Redut Heimpe defect n reduced to the * longitu hetcrog cytogcr __ stantly change' primar The several inhibit tumor systerr The h again: a recr *A jyndn tO i hxvc i betw . NK ceil* and lymphocyte ADCC in preicukemia 243 doubts have lately been cast on the role of CTL in immune defence against cancer, NK cells arc now, based on extensive data, thought to play an important role in the immune surveillance against early neoplastic growth in both humans and rodents (for a recent review, see Hcrberman [10]). This is perhaps best exemplified by investigations in patients with the Chediak-Higashi immunodeficiency (I, 25J, who often die in malignant lymphomaious states, and in whom a low NK activity (and ADCC) against tumor cells is found [9, 25].* Earlier observations comparing NK activity in normal and tumor-bearing individuals have shown a significant decrease in NK activity in tumor patients, the extent of which could be positively correlated to the degree of dissemination of the disease {7, 14]. An important fact to bear in mind, however, is that earlier studies on NK activity in malignant disease have been concerned with patients already harbouring a malignant growth. In contrast, patients with PPS provide an opportunity to test the immunological capacity in individuals with a potentially malignant condition [16, 27], a situation where investigations would seem to yield more information concerning the early immunity against cancer. These considerations have led us to the present investigation of NK activity and ADCC in patients with PPS. In II of 13 patients with a multiparameter-bascd diagnosis of PPS we found a significantly reduced NK activity.! Several factors indicate that this decrease in NK activity could be due to a low number of circulating effector cells and not to an intrinsic defect of the NK cells. Firstly, the correlation between the c/t ratio and target cell killing was similar in the patients and the normal donors (Fig. 2). Secondly, the addition of monocytes, which could be expected to suppress NK cells [32] did not influence the amount of lysis (Fig. 4). Thirdly,- the decreased NK activity was demonstrable against non-mycloid cell lines as well as myeloid cell lines (Fig. 3) indicating that the observed NK . defect is not selective towards myelogenous cells. Finally, enumeration of cells reactive with a monoclonal antibody (901) specific for NK cells (manuscript in preparation) showed significantly reduced numbers of 901-positive cells. Reduced NK activity in PPS has also recently been demonstrated by Porzsolt and Hcimpel [21] and may be either an effect of a subdinical malignant condition or the NK defect may indeed be the primary advent. Favoring the first hypothesis is the finding of reduced NK activity in tumor-bearing individuals [7,14] which also might be comparable to the-demonstration of cytogenetical abnormalities in PPS (19]. On the other hand longitudinal cytogenetical studies performed in our patients with PPS indicate a certain hcterogeneity_ampng_ these, and several of our patients do not show any major cytogenetical abnormalities (to be "published), whereas* their NK activity is.more con stantly reduced. This might speak in favour of an NK defect preceeding the cytogenetical" change. A prospective study on these patients might give some dues as to what is the primary defect triggering evolution towards leukemic proliferation. The in vivo action of a-IFN against tumor cells is under intensive investigation and several mechanisms for an anti-tumoral effect have been considered including a direct inhibitory effect on tumor cdl multiplication [8] or an alteration in the surface antigens of tumor cells and a subsequent increase in the susceptibility to the action of the immune system. Finally, a direct enhancement of anti-tumor host responses could be envisaged. The latter two possibilities require the existence of a functioning host immune defence against tumors. The in vitro action of interferon on NK cells is thought to be due partly to a recruitment of kiiler cells [28] and partly to an increased recycling capacity of the NK A similar impaired NK activity has been obterved by Purtilo in patient* with Xlinked lymphoproliicraiivc lyndrome (24J. tOur control! were not age-matched (younger m ein age for the control*). However. P ro ar>d Baines (23] have recently ihown that aged individual* exhibit higher NK activity than the young. Therefore the companion between our group* would tend to weaken the sLaiiitical tignificancc of the NK differences. 2*6 GtTTE Kernorue et ai. cells [3lj. in vitro Assays for NK activity in our patients with PPS with the addition of a-IFN resulted in augmentation of the cytotoxicity to the same extent and following the same kinetics as for normal controls, but overall cytotoxicity remained subnormal in 12 patients. From this it may be concluded that the number of circulating NK ceils is reduced in PPS, whereas the number of "pre-NK" cells, the cells recruitable by interferon, or their activity s normal. Both myeloid and lymphoid ceils can act as effector cells in the ADCC assay, but lymphocytes seem to be more potent than myeloid cells in the lysis of sensitized tumor cells (17]. Even though this reaction is of importance in (he defense against a variety of microorganisms, its in vivo relevance in tumor immunology is not yet resolved. It is interesting that the ADCC against the P815 mastocytoma cell line was normal (Fig. 5 and Tabic 2) while the NK activity was decreased since these functions arc generally assumed to be exerted by the same lymphoid subset. This dissociation may seem surprising but it is in accordance with the hypothesis that the lytic mechanism in the NK and ADCC reactions are different. In conclusion, our studies, have shown that peripheral blood lymphocytes from preleukemia patients have a severely depressed NK activity, probably due to the low number of circulating NK cells in the blood. Longitudinal studies are now in progress to further elucidate this finding through measurements of peripheral blood and bone marrow NK functions of purified lymphocyte subsets against a variety of myeloid target ceils and against autologous bone marrow cells. REFERENCES 1. A rgvle J. C .. Kieldsberg C. R.. Martt J.. Shiceoka A. O. & Hill K. R. (1982) T-cdl lymphoma and the Chcdiak-Higashi syndrome. Blood 60, 672. ' 2. Bennett J. M., C atovsxy D., Flanorih G .| Galton D. A. G., Gralnick H. R. it Sultan.C. (1982) Pro posals for the classification o f the mydodyxplastic syndromes. Be. J. Haemal. 51, 189. 3. Botum A. (1968) Separation of leukocytes from blood and bone marrow. Scand. J. din. Lab. Invest. 21 (suppl 97). ; 4. Cavallo G.. Gkjvarelu M.. Forni G. A Landolfo S. (1978) Is antibody dependent cellular cytotoxicity an Important mechanism of resistance to tumors in vivo? Immunochcnustry 15, SOI. 5. C lemmesen L.. Hansen M. M .. Mortensen L., T hiede T . A Vioeraek Aa. (1981) The applicability o f simple blood and bone marrow variables in prediction o f acute leukaemia to patients with acquired bi- or pancytopenia. Scand. J. Haemal. 27, 193. 6. E inhorn S.. Blomgren H. A Strand H. (1978) Interferon and spontaneous cytotoxicity in man. (. Enhancement o f the spontaneous cytotoxicity o f peripheral lymphocytes by human leukocyte interferon. Int, J. Cancer 22, *05. 7. Forbes J. T.. Greco A. F. A Ouw a m R. K, (1981) Human natural cell-mediated cytotoxicity. II. Levels to neoplastic disease. Cancer Im mun. Immunother. I I , 147. 8. G resser L., Maury- C. A Brouty-Bote D. (1972) Mechanism o f the antitumor effect o f interferon in mice. -----Nature, Lond. 239, 167.--------------------------------------:----------------------------------------------------------------------- 9. Haliotis T .t Roocr l . %Klein M ,, O rtaldo J., Fauci A. S, & Herrerman R. B. (1980) Chdiak-Higash! gene in humans. (. Impairment o f natural-killer function. J. exp. Med, 151, 1039. 10. HeaBEaMAN R. B.. Ed. (1982) N K Ceils and Other Natural Effector Cells. Academic Press, New York. 11. Heron L, Hoklano M .. Mocler-Larsen A. A Berg K. (1979) The effect ofinicrfcron on lympbocyte- mediated effector cell functions: selective enhancement o f natural killc cells. Ceil Immun. 42, 183. 12. Hoklano P. A Ellegaard J. (1981) Immunological studies in chronic lymphocyte leukemia. II. Natural killer- and antibody dependent cellular cytotoxicity potentials of malignant and non-malignant lymphocyte subsets and the effect of a-interferon. Leukemia Res. 5, 349. 13. Hoklano P.. Ellegaard J. & Heron 1. (1980) ImmunomoduUtion by Corynebocterium parvum in normal humans. J, Immun. 124, 2180. 14. Kadish A. S., Doyle A. T., Steinhauer E. H. A G hossein N. A. (1981) Natural cytotoxicity and interferon production in human cancer: deficient natural killer activity and normal interferon production in patients with advanced disease. J. Immun. 127, 1817, 13. KraCballE K.. Ellegaard J. & Herein T. (1980a) Antibody-dependent monocyte-mediated cytotoxicity, holation and identification by esterase naming of human monocytes detached from monolayers by lower ing the temperature. Scand. J. Haemal. 14. 399. 16. Linman J. W. a Saarni M. I. (1974) The prelcukemic syndrome. Semin. Hemat, 11, 93. 17. MacDonalo H. R.. Bonnard G. D.. Sordat B. A Zawoonik S. A. (1975) Antibody-dependent cell- mediated cytotoxicity: heterogeneity o f effector cells in human peripheral blood. Scand. J. Immun. 4, 487. 18. Marx J. L. (1980) Natural killer cells helps defend the body. Science 110. 624. 19. Panani A., the G-ban* * 20. Pierre R. 21. Porzsolt 22. P rather < cellular cyt Inst. 62. L 23. P ross H. I normal cyt 24. Purtilo D agammagt' 105, 119. 25. Roocr J. ( Fauci A. 26. Rotkstejn 27. Saarni M. Am. J. Me 28. SakselaE. via recruitt 29. Tood R. F monodona 30. T rinchieri derived or antagonist 31. Ullrerg h exp, Med. 32. de Vries J cytotoxic el NK edit and lymphocyte ADCC in prdeukemia 247 19. Panani A.t Papayannis A. G., Kraxu K. A Gadh:as C. (1977) Cytogenetic studies In prdcukacmia using the G-banding technique. Scarni. J. immuti. II, 301. 20. Pierre R. V. (1974) Prdeukcmic states. Semin. Hemal. II , 73. 21. Poacsocr F. A HEiMrCL H. (1982) Impaired T-cell and NK-cell function. Blut 45, 243. 22. Prather O. S., Geller R. W. A Lausch R. N. (1979) Inability of antiserum active in antibody-dependent cdlular cytotoxicity and armine tests to protect against simian virus 40 tumor cdl challenge. J. natn. Cancef Inst. 42, 1273. 23. Paces H. F. A Baines M. G. (1982) Studies of human natural killer celli. 1. In rivo parameters affecting normal cytotoxic functions. Ini. J. Cancer 29, 383. 24. Portilo D. T. (1981) Linked lymphoprolifrative syndrome. An immunodeficiency disorder with acquired agammaglobulinemia, fatal Infectious mononucleosis, or malignant lymphoma. Arch. Path. Lob. Med. 105,119. 25. Rooet I. C.. Haliotis T., Klein M., Koaec S., J ett J. R., OaTALoo J.. Herberman R. B.t Katz P. A Fauci A. S. (1980) A new immunodefidency involving NK cdls. Nature, Land. 284, 553. 2d. Rotketein G. A Wintrobe M. M. (1975) Prdcukcmia. A dv. int. Med. 39, 20. 27. SAAgNt M. I. A Linman J. W. (1973) Prckukemia. The hematologic syndrome proceeding acute leukemia. A m i J. Med. 55, 38. 28. Saxsela E., Timonen T. A Canteu. K. (1979) Human natural killer cdl activity is augmented by interferon j via recruitment of "Pre-NK" cdls. Satnd. J. Immun. 10, 257. f 29. Tooo R. F-, Nadler L. M. A Schlossman S. F. (1981) Antigens on human monocytes identified by monodonal antibodies. J. Im mun. 128, 1435. 30. Tsinchieri G. A Santoli D. (1978) Anti-viral activity induced by culturing lymphocytes with tumorderived or virus-transformed cdls. Enhancement of human natural kilter odi activity by interferon and antagonistic inhibition of susceptibility of target cdls to lysis. J. exp. Med. 147, 1314. 31. Ullberg M. A Jondal M. (1981) Recyding and target binding capacity of human natural killer cdls. J. exp. Med. 153, 615. I 32. de Vries J. E.. Mendelsohn J. A Sont W. S. (1980) Requirement for monocytes in the spontaneous j cytotoxic effects of human lymphocytes against non-lymphoid targets. Nature, Lend. 285, 574. ( i' i f: I < I i I I1i t I I