CELLULAR IMMUNOLOGY 118,448-457 ( 1989)

Lymphokine-Activated Killer Cells in Rats: Generation of Natural Killer Cells and Lymphokine-Activated Killer Cells from Bone

Marrow Progenitor Cells’

MARIANASARNEVA,~NIKOLAL.VUJANOVIC,~ M.R.M. VAN DEN BRINK, RONALD B. HERBERMAN, AND JOHN C. HISERODT

Pittsburgh Cancer Institute and Departments of Pathology and Medicine, University ofpittsburgh. Pittsburgh, Pennsylvania 15213

Received August 18, 1988; accepted October 27, 1988

The coculture of rat bone marrow cells with recombinant interleukin-2 induced the genera- tion of cells mediating natural killer (NK) activity and subsequent lymphokine-activated killer (LAK) activity depending upon the dose of IL-2 and time of culture. NK activity was detected as early as 4 to 5 days after the addition of IL-2 and could be evoked with as little as 5 to 50 U/ ml. The induced NK cells had large granular lymphocyte (LGL) morphology and expressed OX8 and asialo GM, surface markers but did not express OX19 or W3/25 markers. LAK activity was detected only after 5 days of culture, and required above 100 U/ml IL-2. Cells mediating LAK activity also expressed OX8 and asialo GM, but not OX 19. The generation of detectable NK and subsequent LAK activity was due to induction of early progenitor cells and not contaminating mature LGL/NK cells within the bone marrow population since of removal of such mature NK cells with L-leucine methyl ester (L-LME) did not aliect the subsequent generation of either activity. Moreover, the removal of actively dividing cells as well as mature NK cells from the bone marrow by treatment with 5-fluorouracil(5FU) in vivo enriched the remaining bone mar- row population for both NK and LAK progenitor cells. The phenotype of the L-LME- and 5- FU- resistant NK and LAK progenitor cells within populations of bone marrow was determined by antibody plus complement depletion analysis. Although treatment of normal bone marrow with anti-asialo GM, + C reduced the induction of NK and LAK activity in 5-day cultures, treatment of 5-FU marrow with anti-asialo GM, + C did not affect either activity. Treatment with a pan-T cell antibody + C did not affect the development of NK or LAK activity under any conditions. Thus, the 5-FU-resistant NK/LAK progenitors were asialo GM, negative but became asialo GM: after induction by IG2. Finally, evidence that bone marrow-derived LAK cells were generated directly from the IL-2-induced NK cells was obtained by treating the IL-2- induced LGL/NK cells with L-LME. This eliminated the induced NK activity as well as the capacity of the remaining bone marrow cells to generate LAK activity. Thus, in rat bone mar- row, NK cells appear to be derived from asialo GM;, OX 19-, nongranular, 5-FU- and L-LME- resistant progenitor cells which can be induced to generate LGL with NK activity with low levels of IL-2. The IL-Z-induced NK cells became asialo GM:, 0X8+, 0X19- and have a granular

’ Supported in part by Grants CA43765, HL37638, and CA47087 from the National Institutes of Health. 2 Dr. Sameva was on sabbatical leave from the Research Institute of Pharmacology and Pharmacy,

Bulgarian Medical Academy, Department of Oncopharmacology, Sofia, 1156 Bulgaria. 3 Dr. Vujanovic was on sabbatical leave from the Institute of Oncology and Radiology, Pasterova 14,

11000 Belgrade, Yugoslavia.

448

0008-8749/89$3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

GENERATION OF NK AND LAK CELLS FROM BONE MARROW CELLS 449

morphology. Moreover, these IL-2-induced NK cells can be further activated to mediate LAK activity by higher levels of IL-2. o 1989 Academic press, hc.

INTRODUCTION

The control of the growth and differentiation of natural killer (NIQ4 and lympho- kine-activated killer (LAK) cells by recombinant interleukin-2 (rIL-2) has recently been an area of active investigation ( l-3). Although certain subpopulations of T cells are capable of mediating LAK activity (non-MHC-restricted killing of a wide variety of target cells (4-6)), numerous studies have now indicated that the major lymphoid cell population within the peripheral blood or spleen to respond to IL-2 and to gener- ate cells with LAK activity are the LGL/NK cells (4- 10). More recently, studies have suggested a precise lineage relationship among direct NK precursor cells (so-called large agranular lymphocytes), active NK cells (LGL), and IL-2-activated NK cells (LAK cells), a relationship controlled by the levels of IL-2 (11, 12).

NK cells are thought to be derived from bone marrow progenitor cells without the absolute requirement for thymic or other peripheral lymphoid organ processing ( 13- 17). In mice, NK progenitor cells are resistant to treatment with 5-fluorouracil (5- FU) ( 16) and have been shown to express some markers associated with mature NK cells (Thy 1) ( 16) but not other NK-associated markers such as asialo GM, or NK 1 ( 16, 17). Other studies have shown that cells with LAK activity can also be generated from populations of bone marrow cells (17, 18) but the lineage relationship between bone marrow-derived NK cells and cells exhibiting LAK activity has not been system- atically examined. Moreover, there have been no reports of studies on the generation and characteristics of NK cells or LAK cells derived from rat bone marrow.

This study was undertaken to analyze the relationships of the early NK progenitor cells in normal or 5-FU-treated rat bone marrow and to determine whether the induc- tion of LGL with NK activity was a prerequisitie for the induction of cells mediating LAK activity and whether these two activities are linked along a common develop- mental pathway.

MATERIALS AND METHODS

Animals. Male Fischer 344 rats (75-100 g) were purchased from Taconic Farms, Inc. (Germantown, NY) and were housed in a specific pathogen-free animal facility at the Pittsburgh Cancer Institute.

Tumor cells. Routinely, the lysis of the NK-resistant mastocytoma, P8 15, was used as an indicator of LAK activity. Other targets included two NK-resistant syngeneic rat tumor cells: MADB 106 (F344 mammary adenocarcinoma) and CRNK- 16 (F344 LGL leukemia). All of these lines were grown in RPM1 1640 medium with 10% FCS and antibiotics. In several cases, fresh tumor explants were used as targets for assess- ing LAK activity including fresh ascites tumor of CRNK- 16 leukemia and fresh solid tumor explants of the MADB 106 adenocarcinoma. The NK-sensitive Moloney vi- rus-induced YAC- 1 lymphoma was used as the indicator of NK activity.

4 Abbreviations used: NK, natural killer; LAK, lymphokine-activated killer; rIL-2, recombinant human interleukin-2; BM, bone marrow; 5-FU, 54uorouracil; LGL, large granular lymphocyte; L-LME, L-leu- tine methyl ester.

450 SARNEVA ET AL.

Antibodies used. A panel of antibodies was used in these studies. These included mouse IgGl mAb’s, OX8 (CD@, OX19 (CD5), W3/25(CD4), OX6 (Ia), OX39 (CD25), and OX4 1 (monocyte/macrophage), all purchased from Accurate Chemical and Sci- entific Corp. (Westburg, NY). Each of these antibodies was used at 1: 100 dilution based on preliminary dose-response titrations. A complement fixing, pan-T cell monoclonal antibody (Rl-3B3) was obtained from Dr. Craig Reynolds (Biological Response Modifiers’ Program, National Cancer Institute, Frederick, MD) and used at 1: 1,000 dilution. Anti-asialo GM, was purchased from Wako Chemical Co. (Dal- las, TX). All second-step reagents were FITC-labeled F(ab’), fragments of goat anti- body against the primary antibody and were purchased from Cappel Laboratories (Malvern, PA).

Flow cytometry. For surface marker analysis, 2 X lo5 mononuclear cells were placed in 12 X 75-mm glass tubes in 0.1 ml of staining buffer (PBS, pH 7.3, 0.1% sodium azide, 2% FCS). Various antibodies were added for 30 min at 4°C. The cells were washed twice and resuspended in FITC-labeled F(ab’)z fragments of anti-IgG of the primary antibody. After 30 min at 4°C the cells were washed twice, resuspended in 1% paraformaldehyde, and analyzed for fluorescence with a FACStar flow cyto- meter (Becton-Dickinson, Mountain View, CA). For cell sorting, the same protocol as above was used except 1 X 1 O7 cells were initially stained prior to analysis. After cell sorting, each population (positive and negative) was reanalyzed by flow cytometry for purity.

Cytotoxicity assay. Cytotoxicity was measured in a standard 4-hr “Cr release mi- crocytotoxicity assay using 96-well, round-bottomed microplates (Costar, Cam- bridge, MA). The target cells were labeled with 100 j&i of Naz5’Cr04 per 2 X lo6 cells, washed, and seeded into the wells at 5 X lo3 cells/well. Suspensions of effector cells were then added to triplicate wells to give various E:T ratios in a final volume of 200 ~1. After an additional incubation of 4 hr at 37°C 100 ~1 of supernatant was removed from each well and was counted in a gamma counter to determine experi- mental release. Spontaneous release was obtained from wells receiving target cells and medium only, and total release was obtained from wells receiving 1% Triton X- 100. The percentage cytotoxicity was calculated by the following formula: Percentage cytotoxicity = 100 X [(experimental release - spontaneous release)/(total release - spontaneous release)].

Lytic units of cytotoxic activity were determined from linear regression curves plot- ted from various E:T ratios (including 40: 1, 20: 1, 10: 1, and 5: 1). In all cases, 1 LU was defined as the number of effector cells required to cause 20% specific “Cr release from 5 X IO3 target cells.

Preparation of bone marrow cells. Bone marrow cells were obtained from the fe- murs and tibias of normal F344 rats by flushing the marrow cavity with RPM1 1640 and then dispersing the marrow cells through a fine wire mesh. Cell yields were rou- tinely 150-200 X lo6 BM cells per rat. 5-fluorouracil was administered to rats by an intraperitoneal injection of 150 mg/kg in HBSS. This dose was chosen based upon previous studies in mice to deplete more mature hematopoeitic elements. BM cells were then harvested 48 hr later. After harvesting, BM cells were washed twice and suspended in complete medium for use. BM cells obtained from normal or 5-FU- treated rats were cultured in complete medium (RPM1 1640, supplemented with 5% fetal calf serum, 5 X lOA A4 2-mercaptoethanol, streptomycin/penicillin, rgluta- mine, and recombinant human IL-2). The IL-2 was kindly provided by Cetus Corp.

GENERATION OF NK AND LAK CELLS FROM BONE MARROW CELLS 451

(Emeryville, CA) and contained 1.25 X lo6 U of IL-2/mg protein as determined by [3H]thymidine incorporation into CTLL-2 cells.

Antibody and complement depletion studies. BM cells (lO’/ml) were incubated in RPM1 1640 with various antibodies for 30 min at room temperature. Rabbit comple- ment (C) (Low-Tox, Cedarlane, NY) was then added ( 1: 15 final dilution) for an addi- tional 60 min at 37°C. The cells were then washed twice and suspended as to viable cell number in complete medium.

Treatment with L-leucine methyl ester (L-LME). L-LME was prepared and used according to Shau and Golub (11). L-LME (Sigma, St. Louis, MO) was freshly pre- pared in RPM1 1640, adjusted to pH 7.3, and filter sterilized prior to use. Cells to be treated were incubated in 40 mA4 L-LME for 40 minutes at 37°C and washed three times prior to use.

RESULTS

Generation of NK and LAK activities from normal rat bone marrow. Freshly har- vested rat bone marrow contains essentially no detectable levels of NK or LAK activ- ity. However, the addition of rIL-2 to populations of normal F344 bone marrow cells induced the generation of cytotoxic cells mediating NK activity (killing of YAC-1 but not P8 15 target cells) or LAK activity (killing of both target cells) in a dose- and time-dependent manner (Fig. 1). NK activity could be detected as early as 3 to 5 days after the addition of rIL-2 and could be elicited with as low as 5 U/ml. In the presence of 50 U/ml, appreciable levels of cytotoxicity against YAC- 1 cells was noted at 5 days while little or no killing was seen using P8 15, MADB 106, or CRNK- 16 target cells, even with E:T ratios up to 100 to 1. The generation of NK activity correlated with the induction of cells with granular morphology (Fig. 2) and elimination of the IL-2- induced granular cells by treatment with L-LME eliminated the IL-Zinduced NK activity (Table 1). The generation of LGL with NK activity from populations of BM cells did not occur as a consequence of contaminating mature LGL/NK cells within the bone marrow population since the treatment of fresh bone marrow with L-LME (to eliminate possible contaminating mature LGL/NK cells) prior to culture in IL-2 did not influence the induction of new LGL/NK cells from the progenitor cells (Fig. 1, Table 1). Cell sorting analysis (Table 2) indicated that the IL-Zinduced NK cells generated from L-LME-treated bone marrow expressed OX8 and asialo GM, but not OX 19 surface markers which is a characteristic phenotype of mature rat NK cells ( 19, 20). The data in Fig. 1 and Table 3 also show that when BM cells are cultured in the presence of 500 U/ml IL-2, LAK activity (killing of P8 15 and MADB 106 target cells) was detected by Days 5 to 7. Again, treatment of the fresh BM samples with L-LME prior to culture in 500 U/ml IL-2 did not affect the subsequent generation of LAK activity.

NK and LAKprecursor cells are resistant to treatment with 5-fluorouracil. 5-FU is a cell cycle-specific cytotoxic drug which effectively eliminates proliferating cells (2 1, 16). 5-FU will eliminate more mature BM cells at intermediate or advanced stages of maturation while sparing, for the most part, the less actively proliferating stem cells. To determine whether NK progenitor cells in rat bone marrow were resistant to treat- ment with 5-FU, we administered 150 mg/kg of 5-FU, ip, harvested the BM 48 hr later, and tested the remaining bone marrow cells for their response to IL-2. This dose of 5-FU was chosen based on previous studies in mice in which 150 mg/kg was shown

452 SARNEVA ET AL.

UNTREATED

401 5 U/ml

60

I

50 U/ml

60 500 U/ml

L- LME TREATED

0 3 5 7 9

5 U/ml

50 U/ml

500 U/ml

DAYS OF CULTURE

FIG. 1. Kinetics kd dose-response relationship of the generation of NK and LAK activity from normal or L-LME-treated F344 rat BM cells in response to rIL-2.0-0, YAC-1; O-O, P8 15.

to eliminate more mature hematopoietic elements (2 1). The data in Table 4 indicate that while treatment of F344 rats with 5-FU depletes marrow cellularity within 48 hr, there was an enhancement of the remaining S-FU-resistant cells to generate NK activity in response to IL-2 (50 U/ml). The generations of LAK activity (500 U/ml) was also noted in 5-FU-treated and control animals, indicating that 5-FU treatment preserves both NK and LAK progenitor cells.

Surface phenotype of the NK progenitor cells. Using complement-mediated deple- tion we studied the surface phenotype of the IL-2 responsive progenitor cells within populations of normal or 5-F&treated rat bone marrow. As shown in Table 5, deple- tion of cells expressing the asialo GM1 surface marker (-20% of all BM cells) almost completely eliminated the development of cells with NK activity in response to 50 U/ml rIL-2. In contrast, depletion of mature T cells (~4% of all BM cells) did not effect the generation of LGL/NK cells. However, elimination of asialo GM: cells from 5-FU-treated BM did not eliminate the generation of NK cells from the residual

GENERATION OF NK AND LAK CELLS FROM BONE MARROW CELLS 453

- 10

0 1 3 5 7 9

DAYS IN CULTURE

FIG. 2. Generation of LGL and cytolytic activity from L-LME-treated rat BM in response to rIL-2. O-O, YAC-1; o-0, P815.

BM population. Thus, the surface phenotype of the 5-FU-resistant IL-2 responsive progenitor cell within rat bone marrow is asialo GM; and OX 19-.

Generation of lymphokine-activated killer cells from bone marrow-derived LGL/ NK cells. Our data indicate that when bone marrow was cultured under conditions of 50 U/ml IL-2, cells with LGL morphology and NK activity were induced but cells with LAK activity were not induced even after extended culture periods. However, in the presence of 500 U/ml IL-2, cells with LAK activity were observed after 5 to 6 days of culture.

Experiments were done, therefore, to determine if there was a relationship between the cells mediating LAK activity to the cells mediating NK activity. One such experi-

TABLE 1

Effect of L-LME on the Generation or Expression of NK Activity from Normal Rat Bone Marrow Cultured in IL-2

% Cytotoxicity (*SD)

L-LME treatment

Day 0 Day 6

YAC-1 P815 %

4o:l LU/lO’ 40: 1 LU/lO’ LGL

- - 35*4 110 3 1: 0.5 -3 24 f 3 + - 31 +I 100 4 * 0.2 <5 21?1 - + 4 f 0.6 <5 1 rt 0.2 <5 0.6 + + 3+1 <5 1 Z!Z 0.3 4 0.2

Note. F344 BM was untreated or treated with 40 mML-LME and cultured for 6 days in 50 U/ml IL-2. The cultured cells were again treated or not with L-LME and tested for cytotoxicity against YAC-1 or P8 15 target cells or analyzed for the presence of granular cells from cytospins stained with May-Gruenwald Giemsa.

454 SARNEVA ET AL.

TABLE 2

Surface Marker Phenotype of Cells Mediating NK Activity Generated from LLME-Treated Rat Bone Marrow

0 Killing

Expt Population Purity

YAC-1 P815

20: 1 LU/lO’ 20: 1 LU/lO’

Unsorted - 32 160 4 <5 0X8+ 96 58 305 10 12 0X8- 99 4 4 0 <5 Unsorted - 26 135 2 <5 0x19+ 92 4 <5 4 (5 0x19- 98 32 155 4 <5 Unsorted - 36 170 5 <5 Asialo GM : 98 48 245 8 10 Asialo GM ; 98 10 2 3 c5

Note. BM cells were treated with L-LME and cultured for 5 days in 50 U/ml rIG2. The cells were then stained with the indicated antibodies, sorted, and tested for cytotoxic activity against the indicated target cells. The percentage of 0X8+, OX 19+, or asialo GM : cells in the unsorted BM preparations was 12,9, and 28%, respectively.

ment is shown in Fig. 3. NK cells were first generated from LME-treated BM (50 U/ml, Day 6). The induced NK cells were then eliminated with LME and the residual (LME-resistant) cells were cultured under conditions which would ordinarily gener- ate LAK cells (e.g., 500 U/ml). Figure 3 shows that LME depletion of LGL with NK activity at Day 6 almost completely interfered with the further activation of the remaining cells into cells mediating LAK activity when cultured with 500 U/ml IL- 2. Such data indicate that the generation of LAK activity occurs mainly as a conse- quence of further activating the IL-2-induced LGL/NK cells rather than inducing a second population of cells not contained within the LME-sensitive, LGL com- partment.

DISCUSSION

In this report we present data on the nature ofbone marrow-derived NK progenitor cells responsive to IL-2. In addition, we present evidence for a common NK and LAK

TABLE 3

Effect of L-LME Treatment on the Generation of NK and LAK Activity from BM Progenitor Cells

% Killing (40: 1 E:T)

Treatment U/ml rIL-2 YAC-1 P815 MADB106

None 50 42 f 5.5 4 L 0.6 2 f 0.6 L-LME 50 36 -t 2.9 1+1 2 k 0.8 None 500 66 k 4.8 31 e2.9 29 + 1.4 L-LME 500 54k4.1 25 * 1.8 21 kO.9

Note. F344 BM was treated with L-LME (40 mM, 40 min) and cultured in rIL-2 for 6 days.

GENERATION OF NK AND LAK CELLS FROM BONE MARROW CELLS 455

TABLE 4

Effect of in Vivo Treatment with 5-FU on NK and LAK Progenitors

% Cytotoxicity +SD

Treatment U/ml rIL2 YAC-1 P815 MADB106

None 50 26 + 4.2 3 -+ 0.4 2 * 0.4 5-FU 50 36 f 2.1 Ot0.5 0 + 0.1 None 500 64 zk 2.6 32 z!c 5.0 21 * 1.9 5-m 500 51 k6.3 21 + 1.6 16kO.8

Note. Rats received a single ip injection of 5-FU ( 150 m&kg) and 48 hr later, BM ceils were harvested and cultured for 6 days in 50 or 500 U/ml rIL-2. Treatment with 5-FU reduced marrow cellularity from 8 X IO’ cells/femur in control rats to 9 X lo6 cells/femur in 5-FU-treated rats. The results are representative of four experiments and expressed as cytotoxicity at an E:T ratio of 40: I.

progenitor cell and suggest these activities are linked along a common differentiation/ activation pathway. The addition of recombinant IL-2 to cultures of normal rat bone marrow cells induced the generation of cytotoxic cells expressing NK activity in a time- and dose-dependent manner. Cytolytic activity was elicited as early as 3-5 days after the addition of 50 U/ml IL-2 but NK activity could also be induced (albeit to a lesser extent) with 5 U/ml IL-2. The cells generated in these BM cultures had charac- teristics virtually identical to those previously reported for rat NK cells: (1) the cyto- lytic activity was associated with LGL and was directed against target cells typically considered NK-sensitive; (2) the cytolytic activity was associated with cells expressing asialo GM1 and OX8 surface structures but not OX19 antigens which is consistent with the marker phenotype of mature rat NK cells (19,20); and (3) evidence that the

TABLE 5

Effects of Treating Normal or 5-FU Bone Marrow with Various Antibodies plus C on the Generation of NK Activity in the Presence of rIL-2

% Cytotoxicity

Treatment of rats

Treatment of cells prior to culture in IL-2

YAC- 1 P815

20: 1 LU/lO’ 20: 1 LU/lO’

None None/IL-2 48 305 6 15 C/IL-2 42 280 3 15 Anti-asialo GM, + C/IL2 9 10 0 <5 Anti-T cell + C/IL-2 46 285 1 <5

5-FU None/IL-2 58 340 8 10 C/IL-2 55 325 6 15 Anti-asialo GM, + C/IL2 59 210 I 15 Anti-T cell + C/IL-2 51 295 7 15

Note. BM from normal or 5-FU-treated F344 rats (150 mg/kg, 48 hr) was obtained and treated with antibodies plus C, washed, and cultured for 5 days with 50 U/ml rIL2. The cells were collected, washed, and tested for cytolytic activity against YAC-1 and P8 15 target cells.

456 SARNEVA ET AL.

0 YAC 0 PB15

50.

40 l

z

3 4

u E 20. P

0 6 9 11

DAYS IN CULTURE

FIG. 3. Generation of LAK activity from IL2-induced BM-derived NK cells. LLME-treated BM cells were cultured in 50 U/ml IL-2. After 6 days, the cultures were left untreated or again treated with LLME and the remaining cells cultured in 500 U/ml additional IL-2. At each time point shown, cytotoxicity against YAC- 1 and P8 15 target cells (20: 1 E:T ratio) or the percentage LGL (numbers above points) was determined. At Day 6 the percentage LGL in the BM culture was reduced from 26 to 2 and cytotoxicity against YAC- 1 targets reduced from 16 to 0% after GLME treatment. (O-O) L-LME-treated BM cultured in 50 U/ml IL-2 for 6 days then shifted to 500 U/ml and tested on YAC-1; (0-O) L-LME-treated BM cultured in 50 U/ml IL-2 for 6 days then shifted to 500 U/ml and tested on P8 15; (0 --- 0) L-LME-treated BM cultured in 50 U/ml IL-2 for 6 days then again treated with L-LME and shifted to 500 U/ml and tested on YAC-1; (0 --- a) LLME-treated BM cultured in 50 U/ml IL-2 for 6 days then again treated with G LME and shifted to 500 U/ml IL-2 and tested on P8 15.

ILZinduced NK activity was not due to the growth and differentiation of a small population of contaminating mature LGL/NK cells present within the fresh bone marrow samples was obtained from two studies. First, pretreatment of the bone mar- row with the lysosomotropic amine L-LME (which is toxic to granular cells and is known to effectively destroy mature LGL/NK cells (11, 12, 20)) did not affect the resultant development of cytolytic cells from the IL-2-cultured BM. Second, treat- ment of rats with 5-IW (an agent known to eliminate cycling and more mature BM cells (16, 2 1), including mature LGL/NK cells (16)) also did not suppress the in- duction of new LGL/NK cells after culture in IL-2. Indeed, the development of LGL/NK cells from 5-FU-treated marrow appeared to be somewhat more rapid than from untreated marrow indicating that 5-W treatment may even select or enrich for the more primitive IL-2 responsive NK progenitor cells. A similar observation was also made by Migliorati using mouse models (16). Also, as observed with mouse BM cells, the phenotype of the IL-2 responsive BM cells was different in normal and 5- FU-treated preparations. Treatment of normal BM with anti-asialo GM1 + C signifi- cantly depressed the generation of NK (and LAK) activity from the remaining BM

GENERATION OF NK AND LAK CELLS FROM BONE MARROW CELLS 457

population. In contrast, similar treatment of 5-FU-resistant BM cells did not affect the generation of either NK (or LAK) activity. Thus, 5-FIJ-resistant NK progenitor cells appear to be asialo GM; but become asialo GM; following IL-2 activation.

The relationship of the IL-2-generated LGL/NK cells derived from BM progenitors to cells mediating more efficient and broader non-MHC-restricted cytotoxicity (LAK activity) was also investigated. Except for the requirement of higher levels of IL-2 needed to generate cells with LAK activity, our results could not distinguish any features which would differentiate NK and LAK progenitor cells. Evidence that LAK activity could be generated directly from the IL-2-induced NK cells was obtained by depleting the generated NK cells with L-LME or anti-a&o GM1 plus C. Both of these treatments completely eliminated the IL-2-induced NK cells and also eliminated the development of LAK activity upon the subsequent addition of high levels (500 U/ ml) of IL-2. These results indicate that both NK and LAK progenitor cells are con- tained within a common BM population and that their cytolytic activities appear to be sequentially related along a common differentiation and/or activation pathway which is controlled by IL-2.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the excellent secretarial assistance of Ms. Catherine Fekete and Ms. Nancy Harter and the superb technical help of Michael W. Olszowy. We are also grateful to Dr. Raoul R. Salup for his support and helpful suggestions.

REFERENCES

I. Trinchieri, G., and Perussia, B., Lab Invest. 50,489, 1984. 2. Herberman, R. B., and Ortaldo, J. O., Science 214,24, 198 1. 3. Rosenberg, S. A., and Lotze, M. T., Annu. Rev. Immunol. 4,68 1, 1986. 4. Phillips, J. H., and Lanier, L. L., J. Exp. Med. 164,8 14, 1986.

5. Ortaldo, J. O., Mason, A., and Overton, R., J. Exp. Med. 165, 1193, 1986. 6. Lanier, L. L., and Phillips, J. H., Ann. Inst. Pasteur/Immunol. 139,466, 1988.

7. Itoh, K., Tilten, B., Kumagai, K., and Balch, C. M., J. Zmmunol. 134,802, 1985. 8. Salup, R. R., Mathieson, B. J., and Wiltrout, R. H., J. Zmmunol. 138,3635, 1987. 9. Vujanovic, N. L., Herbennan, R. B., Maghazachi, A. A., and Hiserodt, J. C., J. Exp. Med. 167, 15,

1988. 10. Herberman, R. B., Hiserodt, J. C., Vujanovic, N. L. et al. Immunol. Today& 178, 1987. I 1. Shau, H., and Golub, S., J. Zmmunol. 134, 1136, 1985. 12. Maghazachi, A. A., Vujanovic, N. L., Herberman, R. B., and Hiserodt, J. C., J. Immunol. 140,2846,

1988. 13. Hacket, J., and Kumar V., J. Zmmunol. 134,373 1, 1985. 14. Koo, G. C., Dumont, F. J., Tutt, M., Hackett, J., and Kumar, V., J. Immunol. 137,3742, 1986.

15. Pollack, S. B., In “Immunobiology of Natural Killer Cells” (E. Lotzova and R. B. Herberman, Eds.). Vol. 1, p. 87. CRC Press, Boca Raton, FL, 1986.

16. Migliorati, G., Cannaril, L., Herberman R. B., Bartocci, A., Stanley, E. R., and Riccardi, C., J. Immu- nol. 138,36 18, 1987.

17. Yung, Y-P, OkumoraK., andMoore, M. A. S.,J. Zmmunol. 134,1462,1985.

18. Koo, G. C., Peppard, J. R., and Lattime, E. C., Cell. Zmmunol. 98, 172, 1986. 19. Reynolds, C. W., Sharrow, S. O., Ortaldo, J. O., and Herberman, R. B., J. Zmmunol. 127,2204,198 1. 20. Vujanovic, N. L., Herberman, R. B., Olszowy, M. W., Reynolds, C. W., and Hiserodt, J. C., Cancer

Rex 48,884, 1988. 2 1. Van Zant, G., J. Exp. Med. 159,679, 1984.