Lysis of human monocytes by lymphokine-activated killer cells

CELLULARIMMUNOLOGY

11&S-65

(1988)

Lysis of Human Monocytes

by Lymphokine-Activated

Killer Cells

JULIE Y. DJEU AND D. KAY BLANCHARD Department ofMedical

Microbiology and Immunology, University of South Florida College ofMedicine, 12901 North 30th Street, Tampa, Florida 33612 Received May 22, 1987; accepted August 30, I987

Human peripheral blood leukocytes (PBL), stimulated in vitro with recombinant human interleukin 2 (IL-2) for 2-7 days, were seento lyse.autologous and allogeneic monocytes in a 4-hr “Cr-release assay. The lymphokine-activated killer (LAK) cells against monocytic cells were selective in that polymorphonuclear leukocytes (PMN) and nonadherent PBLs were not lysed by these cells. Monocytes which had been cultured for 2-7 days served as better targets than uncultured cells. Also, kinetic studies demonstrated parallel activation of cytolytic activity against monocyte targets and FMEX, an natural killer cell-insensitive human melanoma target. Separation of PBLs by discontinuous density centrifugation identified the effector population in the fractions enriched for large granular lymphocytes (LGL). Precursor cells were seen to express CD2, CD1 1, and some CD16 markers, but not CD3, CD4, CD8, CD15, Leu M3, or Leu 7. The effector population alter IL-2 activation retained the phenotype of the precursor cell. These studies indicate that IL-2 can generate LAK cells against monocytic cells, and this cytolytic activity, especially against autologous monocytes, must be taken into account when IL-2 or LAK cells are used for immunodulation in cancer patients. 0 1988 Academic press, IX. INTRODUCTION

The recent availability of recombinant human interleukin 2 (IL-2) coupled with the rapid accumulation of preclinical data which indicates that IL-2 increases or induces functional activity of cytotoxic T cells, natural killer (NK) cells, and lymphokine-activated killer (LAK) cells against tumor targets and virus-infected targets, have pushed this cytokine to the forefront as a potential immunotherapeutic agent for use in the clinic ( l-6). Promising results from animal as well as human experiments have provided the rationale for testing IL-2 not only in cancer patients for mounting of autologous immune responsesagainst their tumors, but also in patients with acquired immunodeficiency syndrome (AIDS) for reconstitution of their immune response against viral and opportunistic infections. In rodent systems,in vivo IL-2 administration has been reported to restore immunodeficiency in nude mice and allograft responsivenessin B rats (7, 8). In addition, IL-2 enhanced the survival rate of FBL-3 lymphoma-bearing mice that were treated with cyclophosphamide and reconstituted with IL-2-dependent T cells (9). In a number of mouse tumor systems,which include sarcomas, carcinomas, and melanomas metastatic to the lung and liver, concurrent IL-2 administration was apparently required for demonstration of significant antitumor effects, even after adoptive transfer of specific IL-2-expanded cytotoxic cells (2, 10-14). In normal mice, in vivo IL-2 administration was found to enhance NK cell 55 0008-8749188$3.00 Copyright 8 1988 by Academic Press,Inc. All rights of reproduction in any form reserved.

56

DJEU AND BLANCHARD

function ( 15), which is believed to play an important role in immunosurveillance as well asin regulation of a cascadeof immune responsesthrough production of relevant cytokines ( 16-2 1). High-dose IL-2 therapy alone could also cause the regression of established tumors, as has been reported in B- 16 melanoma-bearing mice (22). In the human system, IL2 has been demonstrated to induce the generation of LAK cells against fresh tumor targets which were normally resistant to lysis by peripheral blood leukocytes (PBL) (3, 23). Moreover, IL-2 could readily reconstitute NK activity against K562 tumor cells and cytotoxic activity against cytomegalovirus-infected targets in PBL of AIDS patients (5). These findings have led to an explosion of interest in the clinical use of IL-2. Preliminary clinical trials with IL-2 to determine its toxicity, maximum tolerated dose, and pharmacokinetics, however, have produced surprising results (24-26). Instead of the expected expansion of IL-Zresponsive lymphocytes, the opposite effect was observed in the peripheral blood of treated patients. With a single bolus intravenous (iv) inoculation of IL-2, a sharp decline in circulating lymphoid cell count was noted as early as 15 min after IL-2 administration, with a nadir at 4 hr and a gradual return to normal levels at 24-48 hr (24,25). With continuous iv infusion, a marked decreasewas also noted, with the nadir at Day 3 and a return to normal at Day 7, followed by a rise beyond normal levels (26). The decreasein lymphoid cell number was usually accompanied by clinical symptoms of toxicity such as chills, fever, malaise, headaches, nausea, and in casesof high IL-2 doses,diarrhea and vomitting (2427). Anemia was also associated with LAK plus IL-2 therapy, but a rebound was usually noted within 24-48 hr after cessation of IL-2 administration (28). The loss of circulating lymphoid cells during IL-2 treatment may be due to the redistribution of these cells to other compartments of the body. In fact, intraperitoneal treatment of patients with IL-2 resulted in a 42- to 6600-fold increase in mononuclear cells in the peritoneal exudate cells, with an apparent decreasein monocytes (29). Alternatively, cell loss could be caused by the actual destruction of lymphoid cells by some unknown mechanism induced by IL-2. To determine if the latter possibility could provide part of the explanation for the leukopenia observed after IL-2 treatment, we induced LAK-type effector cells in normal human PBL with recombinant human IL-2 in vitro and tested their lytic activity against a panel of autologous lymphoid cell targets, i.e., adherent monocytes, nonadherent cells (NAC), and PMN. In addition to the kinetics of lysis of the sensitive targets, we also examined the morphology and phenotype of the IL-2-generated killer cells responsible for lysis of autologous normal targets. MATERIALS AND METHODS Preparation of human mononuclear cells. Leukocyte buffy coats obtained from normal volunteers at the Southwest Florida Blood Bank were diluted 1:2 in phosphate-buffered saline (PBS, M. A. Biologics, Walkersville, MD) and layered on 10 ml of Ficoll-Hypaque solution (Pharmacia, Piscataway, NJ) ( 17). After centrifugation at 400g for 20 min at room temperature, the band of PBL at the interphase was collected and washed twice with PBS. The washed PBL were suspended in RPM1 1640 medium containing 5% heat-inactivated human AB serum (Flow, McLean, VA), 2 mM L-glutamine, penicillin, and streptomycin, 5 mM Hepes, and 5 X 10e5 M 2-mercaptoethanol, and will subsequently be referred to as complete medium.

ANTI-MONOCYTE

LAK

ACTIVITY

57

Media reagents were purchased from GIBCO (Grand Island, NY), and plastic supplies were from Costar (Cambridge, MA). For recovery of PMN, the white layer lying on the surface of the red blood cell pellet after Ficoll centrifugation was collected and diluted with 10 vol of 0.83% ammonium chloride to lyse red cells. The PMN were washed twice in PBS and suspended in complete medium. Cytocentrifuged preparations of PMN stained with Giemsa showed greater than 99% PMN by morphology. Preparation of monocytes. PBL were allowed to adhere to plastic petri dishes for 45 min at 37°C to remove adherent cells. Nonadherent cells (NAC) were recovered by vigorous washing of the dishes with warm medium, and the adherent cells were either cultured for 2-7 days at 37°C with fresh medium or removed by gentle scraping with a rubber policeman after the addition of cold PBS. The adherent cells recovered in this manner were determined to contain lessthan 5% lymphocytes by FACS analysis. The adherent cells which were cultured for 2-7 days were recovered in a similar manner. Typically, lo- 15%of the PBL were adherent cells. Discontinuous Percoll density gradient centrifugation. The separation of LGL from small, mature T cells was accomplished by the use of a discontinuous Percoll density gradient. NAC were further depleted of adherent cells and B cells by incubation on nylon wool columns for 30 min at 37°C. The cells passing through the columns were then placed on a six-step discontinuous Percoll density gradient that varied in 2.5% concentrations from 40 to 52.5% Percoll, as described elsewhere (30). After centrifugation at 55Ogfor 30 min at room temperature, the bands of lymphocytes were collected and examined for LGL morphology on Giemsa-stained cytocentrifuged slides. In this series of experiments, fractions 2 and 3 contained 60-80% LGL and represented lo- 15%of PBL, and are referred to as LGL. Serologic depletion studies. LGL, either before or after incubation with IL-2, were washed twice in plain RPM1 and pelleted in culture tubes at 4 X lo6 cells per treatment (17). A previously determined optimal dilution of monoclonal antibody was added directly to cells in a final volume of 0.1 ml, and incubated for 20 min at 37°C. OKT reagents were purchased from Ortho Pharmaceutical Corp. (Raritan, NJ), and the Leu reagents were from Becton-Dickinson (Sunnyvale, CA). Antibodies to OKT4 (CD4) was used to lyse T helper cells: OKT8 (CD8) lyses T cytotoxic/suppressor cells, OKT3 (CD3) lyses all T cells, OKM 1 (CD 11) lyses NK and monocytes, and OKTl 1 (CD2) is effective against most T cells and NK cells. Anti-Leu 7 and antiLeu 11 (CD 16) lyse fresh NK cells, anti-Leu M 1 (CD 15) lyses monocytes/granulocytes, and anti-Leu M3 lyses monocytes/macrophages. Following antibody treatment, low-tox rabbit complement (Cedarlane, Westbury, NY) was added to cells at 1:10 dilution in a final volume of 1.O ml and incubated for an additional 45 min at 37°C to effect lysis. Cells were examined for viability using trypan blue exclusion, washed twice in RPM1 1640, and then resuspended in complete medium to the volume corresponding to the original cell population (4 X lo6 cells per ml) to avoid enrichment of interferring populations. The E:T ratio that is indicated was calculated from the number of viable cells that were present prior to treatment with antibody plus complement. Immunoafinity column separation of Leu 19-labeled efector cells. Since anti-Leu 19 (obtained from Becton-Dickinson) was not cytolytic for labeled cells in the presence of complement, immunoaffinity columns were used in lieu of serologic depletion. Protein A-Sepharose columns were prepared as described (3 1). IL-2-activated

58

DJEU AND BLANCHARD

LGL were washed twice in PBS, incubated with 0.1 ml of anti&u 19 antibodies per 10’ cells for f hr at RT, and washed twice again in PBS. Labeled cells were then loaded onto Protein A-Sepharose columns with bed volumes of 2 ml. The adherent cells (Leu 19+), which typically comprised 65-85% of the original IL-2-activated LGL, were recovered by agitation of the gel bed in combination with increased flow rates. About 70% of the cells which were loaded onto the columns were recovered in this manner. Activation of effectorcells.NAC were incubated at a concentration of 4 X lo6 cells/ ml and LGL were cultured at 2 X lo6 cells/ml in 5- to lo-ml vol with 100 U of recombinant human IL-2/ml, kindly provided by Hoffman-LaRoche, Inc. (Nutley, NJ). Cells were activated for the indicated time in 25-cm* tissue culture flasks, and were then washed twice in medium and readjusted to the original cell concentration. Measurement of cytotoxicity. A 4-hr 51Cr-releaseassay was used to measure the cytotoxicity of NAC or LGL against autologous or allogeneic leukocytes, and FMEX, which is an NK-insensitive melanoma cell line (32). NAC, PMN, and monocytes were labeled in the protocol described for fresh tumor target cells (33). Briefly, leukocytes were labeled with 400 PCi of Na5’Cr04 (Amersham Corp., Arlington Heights, IL) for 2 hr in 0.5 ml of medium. The cells were then washed twice and incubated an additional hour in 5 ml of medium. FMEX target cells were labeled with 200 &i of “Cr per lo6 cells for only 1 hr prior to assay. Target cells were washed twice more and then added to the effector cells at 5 X lo3 cells per well, resulting in effector:target ratios ranging from 40: 1 to 10:1 in a final volume of 0.2 ml in each well. After 4 hr incubation at 37°C the culture supernatants were harvested by removing 0.1 ml of the fluid to be counted in a gamma counter. Maximum isotope incorporated was determined by counting target cells alone, and spontaneous release was measured by counting supernatants of targets incubated with medium alone. The autologous leukocyte targets exhibited lo-30% spontaneous release during the 4-hr assay and the spontaneous releasefrom FMEX tumor targets was typically 5- 10%.The percentage of specific lysis was calculated by the formula ((experimental cpm - spontaneous cpm)/(maximal cpm incorporated)) X 100%. All determinations were done in triplicate and data are reported as the means. The SEM of all assayswas calculated and was 5% or less. Lytic units (LU) were calculated and are defined as the number of effector cells required to lyse 20% of target cells. All results shown are a representative experiment of three or four that were performed. RESULTS

Induction of LAK Activity against FEMX Target Cells FMEX was chosen as the tumor target since it is insensitive to lysis by fresh NK cells, as has been used as an indicator for LAK activity (32). To determine the kinetics of LAK cell generation under the present culture conditions of activation, NAC were incubated for the indicated period of time in the presence of 100 U of IL-2/ml. NAC were then washed in medium to remove IL-2, and adjusted to the original cell concentration. As shown in Fig. 1, optimal activation of cytolytic activity against FEMX tumor target cells occurred after 3 days incubation with IL-2, with levels of activity gradually decreasing thereafter.

ANTI-MONOCYTE

59

LAK ACTIVITY

c2< Days

of

Incubation

With

IL2

FIG. 1. Kinetics of activation of NAC with IL-2. NAC were incubated with 100 U/ml of IL2 for the indicated time and assayedfor cytolytic activity using “Cr-labeled FEMX tumor cells as targets.

Spec$c Lysis ofAutologous Monocytes by LAK Efector Cells The initial experiments were performed to determine whether IL-2 could induce cytolytic activity in NAC of normal individuals against autologous lymphoid cells, and to define which lymphoid subset was susceptible to such lysis. For these experiments, PBL were obtained from individual donors at several time points to determine the maximal susceptibility of fresh and cultured autologous cells to lysis. NAC, optimally activated for 3-4 days with IL-2, were assessedfor their cytolytic activity against autologous monocytes, PMN, or NAC which had been cultured for O-7 days. As shown in Fig. 2, monocytes were specifically lysed by LAK effector cells, and became increasingly more susceptible to lysis with in vitro culture. Optimal monocyte lysis was observed after 3 days of culture, and this high rate of lysis was maintained up to 7 days of culture. Treatment of the monocyte target cells with anti-CD2 antibodies plus complement to eliminate any contaminating T or NK cells slightly augmented cytolytic activity (data not shown). No activity was detected against fresh or 4-day cultured NAC or PMN, or 7-day cultured NAC. Unlike monocytes, NAC and

0

1

2 Days

3 of

Culture

4 Of Target

5

6

7

Cell8

FIG. 2. Time course of susceptibility of cultured monocytes to LAK-mediated lysis. Monocytes (0), NAC (m), and PMN (A) were cultured for the indicated time and used as target cells for autologous NAC, which had been activated with 100 U/ml of IL-2 for 3 days.

60

DJEU AND BLANCHARD TABLE 1 Fractionation of NAC on Percoll Discontinuous Gradients Specific lysis of (LU/ 10’ cells)

Effector” NAC Fraction 2 Fraction 3 Fraction 4 Fraction 5 Fraction 6

IL-2

FEMX

+ + + + + +

112 19 203 19 165 15 70 5 35 11

8

Monocytes* 12 7 6 33 <2 11 <2 12 <2 <2 nd

Cultured monocytes t2 27 10 85 3 32 12 16 <2 6 <2

DCells collected from each fraction of a six-step Percoll gradient were incubated alone or with 100 U/ml of IL-2 for 3-4 days before being assayedfor cytolytic activity against the indicated targets. * Fresh monocytes or monocytes which had been cultured for 3-4 days at 37’C were labeled with “Cr, and used as targets in a 4-hr “Cr-release assay.

PMN were not susceptible to lysis by LAK cells, even after 4-7 days of in vitro culture. These results indicate that IL-2 induced LAK cells specific against monocytes. The anti-monocyte LAK cells therefore resembled those generated by IL-2 against fresh tumor targets reported by other investigators in that lytic activity (i) was generated against autologous targets, and (ii) was not detected in circulating PBL unless they were cultured with IL-2. Percoll Fractionation of Precursor Cells To further define the precursor cell of antimonocyte cytolytic activity, NAC were passedthrough nylon wool to remove contaminating adherent monocytes and B cells before placing the remaining cells on a six-step discontinuous density Percoll gradient. The cells from the various fractions were then collected and tested against FE!MX tumor cells and autologous fresh or cultured monocytes. A representative experiment from three that were performed is shown in Table 1. The IL-Zgenerated killer cells were detected at the highest level in fraction 2 and in decreasing levels in fractions 3 and 4. The level of antitumor cell activity corresponded with the concentration of large granular lymphocytes (LGL), as well as with the level of antimonocyte activity. Fresh NK cells are also known to fall in fractions 2-3 of the Percoll gradient (30). Therefore, autolytic precursor cells appeared to be of LGL morphology and copurified with NK cells in the Percoll gradient. Lysis ofAutologous and Allogeneic Monocytes To determine if the antimonocyte activity of LAK effector cells were specific for autologous target cells, LGL were purified from separatedonors and cytolytic activity against autologous and allogeneic cultured monocytes was assessed.A representative

ANTLMONOCYTE

61

LAK ACTIVITY

TABLE 2 Lysis of Autologous and Allogeneic Monocytes by LAK Cells Specific lysis of target (LU/ 10’ cells) Donor 1 2

Effector” LGL LGL+IL2 LGL LGL+IL2

FEMX

I-Monocyte b

2-Monocyte

20 231 13 293

8 28 2 60

15 65 8 98

a LGL obtained from Percoll fractions 2 and 3 were cultured alone or with 100 U/ml of IL-2 for 3-4 days prior to being assayedfor cytolytic activity against the indicated targets. b Monocytes obtained from Donors 1 and 2 were cultured for 3-4 days, labeled with “Cr, and used as targets for autologous and allogeneic effector LGL.

experiment from four that were performed is shown in Table 2. Cytolytic activity was generated by IL-2 against FEMX tumor cells and cultured monocyte targets from two separate donors, indicating that lysis of monocytes by activated LGL was not MHC restricted. Phenotype of the Precursor Cell To determine the phenotype of the LAK precursor cell in the LGL population, we treated the LGL from fractions 2-3 with complement plus monoclonal antibodies to CD4/CD8, CD 15, Leu M3, CD3, Leu 7, CD1 6, CD2, or CD1 1. These cells were then incubated with IL-2 and tested for antitumor and antimonocyte activity after optimal activation. Table 3 shows that antibodies to CD4 which lysed T helper cells and CD8 TABLE 3 Phenotype of Effector Precursor Cells in LGL Specific lysis of target (LU/ 10’ cells) Treatment of LGL”

FEMX

Automonocyte b

Allomonocyte

c’ control CD3 CD4/CD8 Leu M3 Leu7 CD16 CD11 CD2

177 153 179 199 165 37 4 2

23 20 28 39 26 10 <2 <2

67 65 81 79 58 9 <2 12

% Viability of LGL’ 99.5 66.7 76.1 94.1 60.9 37.6 48.1 9.4

a LGL recovered from Percoll fractions 2 and 3 were treated with complement alone or with each of the indicated monoclonal antibodies plus complement before being cultured for 3-4 days with 100 U/ml of B-2. bAutologous (auto-) or allogeneic (allo-) monocytes were cultured for 3-4 days prior to being labeled and used as target cells. ‘Number indicates the percentage of viable cells after treatment with complement alone or with each of the indicated antibodies plus complement.

62

DJEU AND BLANCHARD TABLE 4 Phenotype of Activated LGL Specific lysis of target (LU/ 10’ cells)

Treatment of LGL” C control CD3 CD4/CD8 Leu M3 LeUl

CD16 CD11 CD2

FEMX 211 219 231 255 181 141 17 12

Automonocyte” 104 73 86 101 88 39 10 9

Allomonocyte 152 116 120 155 117 58 21 14

% Viability of LGL’ 99.5 65.9 83.5 96.7 82.8 61.6 66.2 40.0

’ LGL recovered from fractions 2 and 3 of Percoll gradient were incubated for 3-4 days at 37’C with 100 U/ml of IL-2 and then treated with complement alone or with each of the indicated monoclonal antibodies plus complement. b Autologous (auto-) or allogeneic (allo-) monocytes were cultured for 3-4 days prior to being labeled and used as target cells. ‘Number indicates the percentage of viable cells after treatment with complement alone or with each of the indicated antibodies plus complement.

which lysed T cytotoxic/suppressor cells had no deleterious effect on the IL-2 generation of effector cells, and treatment with anti-CD3, which lysed all T cells, had only a slight effect on cytolytic activities. Antibodies to CD1 5 which lysed monocytes and granulocytes and Leu M3 which lysed monocytes also had no adverse effect on both activities. These results indicated that the few contaminating T cells and monocytes in the LGL preparation played no part in IL-2 generation of LAK functions. Treatment of LGL with anti-Leu 7 did not inhibit the generation of IL-Zresponsive killer cells, contrary to its ability to partially inhibit NK activity of fresh LGL. Fresh LGL which were treated with anti-Leu 7 plus complement lost about 40% of cytolytic activity against K562 in our hands (data not shown). Antibodies to CD 16, which completely inhibited fresh NK activity (not shown), had only a partial inhibitory effect on IL-2 generation of an&FEMX or anti-monocyte activity. In contrast, treatment with antibodies to CD2 and CD 11 completely blocked the IL-2 generation of cytolytic activity. Additionally, the precursor of antiautologous and antiallogeneic monocyte activity in LGL was similar if not identical.

Phenotypeof the Eflector Cells To determine the phenotype of the effector cell, LGL that had been preincubated with IL-2 for 3-4 days were treated with a panel of monoclonal antibodies plus complement as described earlier. These cells were then tested against FEMX tumor cells, and autologous or allogeneic cultured monocytes. As shown in Table 4, the phenotype of the precursor killer cell was retained and no new markers were induced by IL-2. Since the NKH- 1 or Leu 19 antigen has recently been linked with LAK activity as well as IL-2-dependent CTL clones (34-36), it was used to determine if Leu 19+ cells were responsible for antimonocyte cytotoxicity. LGL were activated with IL-2,

ANTI-MONOCYTE

63

LAK ACTIVITY

TABLE 5 Separation of Leu 19-Labeled LAK Cells on Protein A-Sepharose Immunoaffinity Columns Specific lysis of target (LU/ 10’ cells) Effector celln LGL Leu 19+ Leu 19-

FEMX

Monocyte

Culture monocyte b

3 44 <2

149 351 27

563 1037 108

’ LGL were incubated with 100 U/ml of IL-2 for 3-4 days prior to separation into Leu 19+and Leu 19populations on Protein A-Sepharose columns. b Fresh monocytes or monocytes which had been cultured for 3-4 days were labeled with “Cr and used as target cells.

and then separatedinto Leu 19- and Leu 19+populations using Protein A-Sepharose immunoaffinity column chromatography. As shown in Table 5, Protein A-Sepharose nonadherent cells (Leu 19-), were largely depleted of LAK activity against both tumor and autologous monocyte targets. Furthermore, there was an enrichment of cytolytic activity in the Leu 19+, or adherent, cells. DISCUSSION The present study provides evidence that IL-2 is capable of inducing cytolytic activity in NAC against autologous and allogeneic monocytes. This antimonocyte activity appeared to be selective because PMN and NAC could not be lysed, either when freshly isolated or after up to 7 days of in vitro culture. On the other hand, monocytes became even more susceptible to lysis by IL-2-induced LAK cells after 3-7 days in culture, suggesting that the target recognition structure(s) may be related to maturation or differentiation antigens specific to macrophages. Pretreatment of monocyte target cells with anti-CD2 antibodies plus complement to eliminate contaminating T or NK cells resulted in slightly augmented cytotoxicity, further identifying the target cells in plastic-adherent PBL as monocyte/macrophages (data not shown). LAK-type cells against autologous monocytic cells required a minimum of 2 days of in vitro culture with IL-2 for their induction, similar to LAK cells reported against fresh tumor targets (3,23). They shared another similarity to LAK cells in that activity could be induced against autologous tissue but was absent in PBL until IL-2 stimulation. In that respect, they did not resemble circulating NK cells, which were lytic for K562 tumor cells in the resting state prior to exposure to IL-2 (4,5). The antimonocyte LAK effector cells in the present study phenotypically resembled that reported for LAK-type cells against NK-resistant tumor cells. They were CD2+ and CD 1l+, and negative for other T-cell or monocyte markers, such as CD4, CD8, CD15, CD3, and Leu M3 (3, 35, 37). Additionally, there were no new surface markers induced during activation that were detected using the indicated monoclonal antibodies. The effector cells also appeared to be Leu 19+, which has been most closely linked with LAK activity and IL-2-dependent CTL clones (34-36). While several investigators have reported at least two populations of cells that are Leu 19+,

64

DJEU AND BLANCHARD

i.e., CD3- and CD3+ cells (34, 35), we were unable to determine if one or both subpopulations of Leu 19+cells were cytolytic for monocytes. Sondel et al. (38) recently reported similar findings as ours, in which they described the generation of cytotoxic cells against autologous mononuclear targets by 6-day culture in IL-2. These investigators did not observe autoreactivity in PBL until after activation with IL-2. Becausethe targets used in their study were unseparated PBL, the level of lysis was considerably lower than that observed by us. This low lysis is likely due to the small number of monocytes in PBL that are selectively lysed by LAK cells. The LAK-type cytotoxicity against monocytes was not MHC restricted because both autologous and allogeneic monocytes were destroyed. Monoclonal antibody depletion studies showed that the same population of IL-2-activated LGL that had the capability to lyse autologous monocytes also lysed allogeneic monocytes. Moreover, the cell fractions that lysed monocytes were the only populations that lysed FEMX tumor targets. These results suggestedthat the same effecters may be involved in lysis of autologous/allogeneic monocytes and FEMX tumor cells. The reason for the development of antimonocyte activity in LAK cells is not clear. One possibility is that LAK cells serve in the feedback regulatory mechanism for eliminating monocytes required for interleukin 1 production, and subsequent IL-2 production, when excessIL-2 is present. The destruction of monocytes by LAK cells is probably not the primary factor influencing the disappearance of mononuclear cells from the circulation during the acute phase of IL-2 administration. However, some of the monocytopenia reported in the later phase of IL-2 administration or LAK/IL-2 immunotherapy could be caused by the direct effect of LAK cells on monocytes. Our observation that cultured monocytes were more susceptible to LAK-mediated lysis than uncultured monocytes suggeststhe additional possibility that tissue macrophage destruction by LAK cells may take place during LAK/IL-2 therapy. IL-2-generated killer cells against K562 tumor cells have recently been detected in the lamina propria of patients with colon carcinoma. The phenotype of the anti-K562 killer cells induced by IL-2 in the GI tract was reportedly CD2-, CD3-, Leu 7-, and CD 16+, in both the precursor and the activated states (39), and resembled the phenotype of the antimonocyte killer cells. It is therefore tempting to speculate that destruction of autologous normal monocytes/macrophages may have contributed to GI toxicity that has been reported in IL-2-treated individuals. This type of tissue destruction could easily have gone unnoticed because of the inaccessibility of the cells for in vitro study. Further studies will be necessaryto determine the biological consequencesof elimination of monocytes by LAK cells, and to ascertain if LAK cell lysis of monocytes contributes to certain immunotoxicities of LAK/IL-2 immunotherapy. REFERENCES 1. Taniguchi, T., Matsui, H., Fug&a, T., Takaoka, C., Kashina, N., Yoshimoto, R., and Hamuro, J., Nature [London) 302,305, 1983. 2. Rosenberg, S., .I. Natl. Cancer Inst. 75,595, 1985. 3. Grimm, E. A., Ramsey, K. M., Mazumder, A., Wilson, D. J., Djeu, J. Y., and Rosenberg, S. A., J. Exp. Med. 157,884,1983. 4. Domzig, W., Stadler, B. M., and Herberman, R. B., J. Zmmunol. 130, 1970, 1983.

ANTI-MONOCYTE

65

LAK ACTIVITY

5. Rook, A. H., Masur, H., Lane, H. C., Frederick, W., Kasahara, T., Macher, A. M., Djeu, J. Y., Manischewitz, J. F., Jackson, L., Fauci, A. S., and Quinnan, G. V., J. Clin. Invest. 72,398, 1983. 6. Rook, A. H., Hooks, J. J., Quinnan, G. V., Lane, H. C., Manischewitz, J. F., Macher, A. M., Masur, H., Fauci, A. S., and Djeu, J. Y., J. Zmmunol. 134, 1503, 1985. 7. Stotter, H., Rude, E., and Wagner, H., Eur. J. Zmmunol. 10,719, 1980. 8. Clason, A. E., Duarte, A. J. S., Kupiec-Weglinski, J. W., Williams, J. N., Wang, B. S., Strom, T. B., and Tilney, N. L., J. Zmmunol. 129,252, 1982. 9. Cheever, M. A., Greenberg, P. D., Fefer, A., and Gillis, S., J. Exp. Med. 155,968, 1982. 10. Mule, J. J., Shu, S., and Rosenberg, S. A., J Zmmunol. 135,646, 1985. 11. Lafreniere, R., and Rosenberg, S. A., Cancer Res. 45,3735, 1985. 12. Lafreniere, R., and Rosenberg, S. A., J. Zmmunol. 135,4273, 1985. 13. Ettinghausen, S. E., and Rosenberg, S. A., Cancer Rex 46,2784, 1986. 14. Cheever, M. A., Greenberg, P. D., Irle, C., Thompson, J. A., Urdal, D. L., Mochizuki, D. Y., Henney, C. S., and Gillis, S., J. Zmmunol. 132,2259, 1984. 15. Chang, A. E., Hyatt, C. L., and Rosenberg, S. A., J. Biol. Response Mod. 2,56 1, 1984. 16. Herberman, R. B., “NK Cells and Other Natural Effector Cells.” Academic Press,New York, 1982. 17. Djeu, J. Y., Stocks, N., Zoon, K., Stanton, G. J., Timonen, T., and Herberman, R. B., J. Exp. Med. 156,1222,1982.

18. Kasahara, T., Djeu, J. Y., Dougherty, S. F., and Oppenheim, J. J., J. Zmmunol. 131,2379, 1983. 19. Scala, G., Djeu, J. Y., Allevena, P., Kasahara, T., Ortaldo, J. R., Herberman, R. B., and Oppenheim, J. J., Nature (London) 309,56, 1984. 20. Procopio, A. D. G., Allavena, P., and Ortaldo, J. R., J. Zmmunol. 135,3264, 1985. 21. Burlington, D. B., Djeu, J. Y., Wells, M. A., and Quinnan, G. V., J. Zmmunol. 132,3 154, 1984. 22. Rosenberg, S. A., Mule, J. J., Speiss,P. J., Reichert, C. M., and Schwartz, S., J. Exp. Med. 161, 1169, 1985. 23. Itoh, K., Tilden, A. B., Kamagai, K., and Balch, C. M., J. Zmmunol. 134,802, 1985. 24. Bindon, C., Czemiecki, M., Ruell, P., Edwards, A., McCarthy, W. H., Harris, R., and Hersey, P., Brit. J. Cancer47, 123, 1983. 25. Lotze, M. T., Frana, L. W., Sharrow, S. O., Robb, R. J., and Rosenberg, S. A., J. Zmmunol. 134, 157, 1985. 26. Lotze, M. T., Matory, Y. L., Ettinghausen, S. E., Raynor, A. A., Sharrow, S. A., Seipp, C. A. Y., Custer, M. C., and Rosenberg, S. A., J. Zmmunol. 135,2865, 1985. 27. Rosenberg, S. A., Lotze, M. T., Muul, L. M., Leitman, S., Chang, A. E., Ettinghausen, S. E., Matory, Y. L., Skibber, J. M., Shiloni, E., Vetto, J. T., Seipp, C. A., Simpson, C., and Reichert, C. M., N. Engl. J. Med. 313, 1485, 1985.

28. West, W. H., Tauer, K. W., Yannelli, J. R., Marshall, G. D., Orr, D. W., Thurman, G. B., and Oldham, R. K., N. Engl. J. Med. 316,898, 1987. 29. Lotze, M. T., Custer, M. C., and Rosenberg, S. A., Arch. Surg. 121, 1373, 1986. 30. Timonen, T., and Saksela,E., J Zmmunol. Methods 36,285, 1980. 3 1. Allavena, P., and Ortaldo, J. R., In “Immunobiology of Natural Killer Cells” (E. Lotzova and R. B. Herberman, Eds.), Vol. 1, pp. 2 l-39. CRC Press,Boca Raton, FL, 1986. 32. Lindmo, T., Davies, C., Fodstad, O., and Morgan, A. C., Znt. J. Cancer 34,504, 1985. 33. Rayner, A. A., Grimm, E. A., Lotze, M. T., Wilson, D. J., and Rosenberg, S. A., J. Nat/. Cancer Inst. 75,67, 1985. 34. Phillips, J. H., and Lanier, L. L., J. Exp. Med. 164,8 14, 1986. 35. Ortaldo, J. R., Mason, A., and Overton, R., J. Exp. Med. 164,1193, 1986.

36. Lanier, L. L., Le, A. M., Ding, A., Evans, E. L., Krensky, A. M., Clayberger, C., and Phillips,

J. H., J.

Zmmunol. 138,2019, 1987.

37. Allavena, P., and Ortaldo, J. R., J. Zmmunol. 132,2363, 1984. 38. Sondel, P. M., Hank, J. A., Kohler, P. C., Chen, B. P., Minkoff, D. Z., and Molenda, 137,502, 1986. 39. Hogan, P. G., Hapel, A. J., and Doe, W. F., J. Zmmunol. 135, 1731, 1985.

J. A., J. Zmmunol.