Lymphokine-Activated Killer Cells: Lysis of Fresh Syngeneic Natural Killer-Resistant Murine Tumor Cells by Lymphocytes Cultured in Interleukin 2

[CANCER RESEARCH 44, 1946-1953, May 1984]

Lymphokine-activated Killer Cells: Lysis of Fresh Syngeneic Natural Killer-resistant Murine Tumor Cells by Lymphocytes Cultured in Interleukin 2

Maury Rosenstein, llana Yron, Yael Kaufmann, and Steven A. Rosenberg1

Surgery Branch, Division of Cancer Treatment, National Cancer Institute, NIH, Bethesda, Maryland 20205

ABSTRACT optimal conditions for the production of IL-2 and long-term in vitro growth of both murine and human T-lymphoid cells (17, 27- Normal splenocytes that are cultured in the lymphokine, inter- 32, 37). In human studies, the LAK cells may be induced with leukin 2 (IL-2), for as short as 2 days develop lytic activity for only a 2-day incubation of autologous lymphoid cells in the fresh syngeneic natural killer-resistant tumor cells as well as presence of IL-2. These cells expressed the phenotypic charac natural killer-sensitive YAC cells in a 4-hr 51Cr release assay. teristics of allogeneic CTL based on their susceptibility to lysis Lymphokine-activated killer (LAK) cells do not lyse syngeneic by OKT-3 and OKT-8 monoclonal antibody and complement, fresh lymphocytes but do lyse syngeneic concanavalin A-induced and lysed a variety of fresh autologous human tumor cells as lymphocyte blasts. Lysis is not due to the presence of lectin or well as fresh allogeneic tumor cells, but not normal lymphocytes xenogeneic serum and appears to be an intrinsic property of (8-10). Little information exists, however, as to the presence of lymphocytes activated in IL-2. The activation appears universal LAK cells and their characterization in the mouse (37). In this in that lymphocytes from all strains of mice activated in this paper, we report on the presence of LAK cells in a variety of manner exhibited similar patterns of lysis for fresh tumor target mouse strains, the precursor and effector phenotypes of the cells. murine LAK cells, and the specificity of lysis of these cells. The To characterize the cells responsible for this lysis, we analyzed high levels of lysis by LAK cells of syngeneic NK-resistant tumor the phenotypic expression of surface markers on these cells cells may be of value in studying interactions between lymphoid with depletion techniques using monoclonal antibody and com cells and tumors as well as for adoptive immunotherapy. plement. These studies indicate that the precursor of the LAK cell is Thy-1+ and nonadherent to plastic or nylon wool. Lysis of syngeneic tumor was inhibited when LAK cells were treated with MATERIALS AND METHODS an anti-Thy-1.2, or anti-Lyt-2.2 monoclonal antibody and comple Animals. C57BL/6, BALB/c, DBA/2, AKR, C3H, and B10.D2 mice ment but not with anti-Lyt-1.2 monoclonal antibody and comple were obtained from the Animal Production Colonies of the NIH, Bethesda, ment, indicating that the observed lytic activity was due to a Thy-1+ Lyt-1~2+ cell. Furthermore, LAK cell-mediated lysis could MD. Tumors. Two sarcomas, MCA-102 and MCA-106, were induced by be inhibited by the addition of anti-Lyt-2 or LFA-1 monoclonal the injection of 0.1 ml of 1% 3-methylcholanthrene in sesame oil into the antibody to cytotoxicity assays. Cold target inhibition analysis right hind limb of female C57BL/6 mice as described previously (24). The revealed that the syngeneic tumor cells were lysed by recognition tumors were maintained by repeated passage in syngeneic hosts and of a determinant not present on normal lymphocytes or lympho used in these experiments in the fourth to seventh transplant generation. P-815 (H-2d) was passaged in ascites form in DBA/2 syngeneic mice, cyte blasts. This lysis of fresh solid tumor cells by lymphoid cells and EL-4 (H-26) was passaged in ascites form in C57BL/6 mice. grown in IL-2 may be of value in the study of tumor-host Preparation of Tumor Cell Suspensions. Tumor cell suspensions immunological interactions. The biological significance of tumor were prepared by short trypsinization as described previously (37). In lysis by IL-2-activated cells requires further study. brief, tumors were cut into 5-cu mm fragments in HBSS, washed once with 0.25% trypsin in Dulbeco's phosphate-buffered saline without Ca2+ and Mg2+ (NIH Media Unit), and then trypsinized at room temperature INTRODUCTION for 10 min. The supernatant containing released cells was removed into The ability to generate T-cells with lytic activity for fresh an equal volume of cold RPMI Medium 1640 (Grand Island Biological autologous tumor after specific in vitro sensitization has met Co., Grand Island, NY), penicillin (100 units/ml), and streptomycin (100 with limited success in both murine and human studies. In an Â¿ig/ml).This medium is referred to in this paper as RPMI medium. An attempt to reproducibly generate T-cells with lytic activity to equal volume of fresh trypsin solution was added to the trypsinization flask, and the procedure was repeated once. The pooled cell suspension fresh autologous tumor, we have been studying the nonspecific supernatants were washed 3 times with cold RPMI medium by centrifu- activation of T-lymphoid cells caused by short-term culture in IL- 2.2 gation for 7 min at 100 x g. In order to lyse erythrocytes, cell suspensions were treated with In previous reports from our laboratory, we have described buffered ammonium chloride lysing solution. Preparation of Splenocyte Suspensions. Spleens were removed 1To whom requests for reprints should be addressed, at Surgery Branch, aseptically into HBSS and were passed gently through a 40 mesh screen National Cancer Institute, Building 10, Room 2B42, Bethesda, MD 20205. and through a single layer of 100 mesh nylon. The cells were centrifuged 2 The abbreviations used are: IL-2, interteukin 2; LAK, lymphokine-activated at 100 x g for 7 min, and the pellet was resuspended in a buffered killer; CTL, cytotoxic T-lymphocytes; Con A, concanavalin A; CM, complete me dium; HBSS, Hank's balanced salt solution; NK, natural killer; MCA, methylchol- ammonium chloride solution for 1 min at room temperature to lyse the anthrene-induced sarcoma. RBC. After 1 min, the suspension was diluted in RPMI medium and Received September 22, 1983; accepted December 21,1983. washed 3 times. For the growth of T-cells, the final cell pellet was

1946 CANCER RESEARCH VOL. 44

den, CT) for 30 min at 37Â°; 100 n\ of the 51Cr-labeled MCA-102 tumor suspended in complete RPMI Medium 1640, (Grand Island Biological Co.) supplemented with 0.03% fresh glutamine (NIH Media Unit), 1 UM target cells were then added to the wells. The plates were incubated for an additional 4 hr prior to harvest as described for the 51Cr release assay. sodium pyruvate (Microbiological Associates, Walkersville, MD), 0.1 ITIM nonessential amino acids (Microbiological Associates), 5 x 105 M 2- Role of Lyt-2 and LFA-1 in LAK Effector Cell Lysis. The effect of mercaptoethanol, penicillin (100 units/ml), streptomycin (100 /Â¿g/ml),and anti-Lyt-2 and anti-LFA-1 monoclonal antibodies on cytotoxicity of CTL 10% heat-inactivated fetal calf serum (Grand Island Biological Co.). and LAK was studied by incubating 5x10* effector cells in microtiter Preparation of Con A-stimulated Blast Cells (Con A Blasts). Spie- plates with various dilutions of the following monoclonal antibodies: (a) nocytes (4 x 106) were incubated for 48 hr in 2 ml of CM containing 10 anti-Lyt-2 53.6.7 (15), concentrated 10 times; and (b) anti-LFA-1 H35- ^g of Con A in the wells of Costar 24-well plates (Catalogue No. 3524; 89.9 (25) ammonium sulfate precipitate from ascites, kindly supplied by Dr. P. Golstein and Dr. M. Pierres, Centre d'Immunologie, France. After (Costar, Cambridge, MA). 30-min incubation at 37Â°in 150 n\ of medium, 5 x 103 51Cr-labeled target In Vitro Sensitization. Allogeneic in vitro sensitizations were per formed using conventional techniques (27, 29, 30). In brief, spleen cell cells were added in 50 p\. Plates were spun for 5 min at 500 rpm and suspensions were prepared as described above. Stimulator cells re incubated for 5 hr at 37Â°.Assays were harvested as described above. ceived 2000 rads in a 7-irradiator. Responder cells (4 x 106) and 106 Adherent Cell Depletion. Splenocytes were incubated at 2 x 10'' irradiated stimulator cells were mixed and added to each well of a Costar cells/ml in 45 ml of CM in supine 650-ml tissue culture flask (Costar) at 24-well tissue culture plate in 2 ml of CM. The cells were then incubated 37Â°for 1 hr. Nonadherent cells were removed and incubated for 1 hr at for 4 days at 37Â°, 5% CO2. In experiments where large numbers of 37Â° in a 50-ml centrifuge tube containing 0.8 g nylon wool, and the alloreactive cells were required for antibody and complement depletion, nonadherent cells were harvested and resuspended in CM. 75 x 106 responder cells were admixed with 15 x 106 irradiated stimu lator cells (2000 R) in 90 ml of CM in 250-ml upright culture flasks (Falcon RESULTS Plastics, Oxnard, CA). The cells were incubated at the culture conditions described previously. Cytotoxic Activity and Specificity of LAK. Normal fresh Preparation of IL-2. IL-2, free of lectin, was prepared as described previously (31). In brief, 2 x 109 BALB/c splenocytes were incubated in C57BL/6 lymphocytes or lymphocytes cultured for 5 days in CM 50 ml of CM containing Con A (10 ^g/ml) at 37Â°,5% CO2 for 2 hr. The or IL-2 were tested for lytic activity versus several normal and cells were then washed 3 times with HBSS, and the pellet was resus- tumor targets in a 4-hr chromium release assay (Table 1). Fresh pended in 50 ml of CM. Two ml of the washed cell suspension were C57BL/6 lymphocytes or lymphocytes cultured in complete me placed into individual wells of 24-well Costar tissue culture plates and dium for 5 days demonstrated 20 and 8% lysis, respectively, for incubated for an additional 24 hr. The supernatant was collected and the YAC, NK-sensitive target, but not for any of the other targets treated as described previously. IL-2 prepared in this manner was greater tested. Lymphocytes activated in IL-2 for 5 days manifested than 95% Con A-free (31). Highly purified human IL-2 produced by the significant lysis for the fresh, NK-resistant, syngeneic MCA-102, high producer Jurkat clone and purified by passage over an anti-IL-2 MCA-106, and EL-4 lymphoma as well as the YAC target. No affinity column was kindly supplied by Dr. R. Robb, E. l. duPont, lysis of fresh C57BL/6 lymphocytes was observed although good Glenolden, PA. cytotoxicity was found against C57BL/6 syngeneic Con A blasts. Growth and Activation of LAK. For the activation of LAK cells, splenocytes were prepared at a concentration of 4 x 106 viable cells/ml. These reults indicated that IL-2 will activate fresh lymphocytes One ml of the cell suspensions and 1 ml of IL-2 were added to each well to kill NK-resistant tumor and blast target cells. of a Costar tissue culture plate. After 2 to 7 days at 37Â°with 5% CO2, Kinetics of LAK Activation. Normal C57BL/6 lymphocytes cells were harvested, counted in 0.25% trypan blue, and resuspended were activated in IL-2 or CM and tested for in vitro cytotoxicity either in complete medium or in RPMI medium for cytotoxicity testing. at 100:1 effectortarget ratios versus 3 syngeneic fresh-tumor Splenocytes used for effector and precursor cell experiments were target as well as C57BL/6 fresh lymphocytes and Con A blasts lymphokine-activated for 3 to 4 days at a concentration of 2 x 106 cells/ in a 4-hr chromium release assay (Chart 1). Appreciable lytic ml in either 50% IL-2 or at various dilutions of Jurkat IL-2. Cells were activity for tumor targets was observed by 2-day culture in IL-2 cultured in a final volume of 2 ml in 24-well Costar plates, 18 to 20 ml in with peak reactivity on Day 4 of culture. LAK cells at no time 30-ml supine culture flasks, or in 150 ml in 750-ml supine culture flasks. lysed fresh C57BL/6 lymphocyte; however, substantial lysis was Depletion with Antibody and Complement. Fresh C57BL/6 lympho cytes were adjusted to a concentration of 107 cells/ml in cytotoxicity seen for Con A blasts. Fresh C57BL/6 lymphocyte or lympho medium composed of RPMI 1640, 0.3% bovine serum albumin, and 25 cytes cultured in CM for up to 7 days at no time demonstrated HIM4-(2-hydroxyethyl)-1-piperazineethanesulfonicacid buffer, containing significant lysis for any of the targets tested. monoclonal antibody at the desired final concentration. Anti-Thy-1.2, IL-2 as Stimulus for the Generation of LAK Cells. The lytic anti-Lyt-2.2, and Lyt-1.2 were used at a 1:500,1:100, and 1:200 dilution, capacity of cells cultured in supernatants containing IL-2 was respectively (New England Nuclear, Boston, MA). Cells were incubated due only to IL-2 and not Con A, fetal calf serum, or other for 30 min at 37Â°with intermittent agitation, were centrifuged, and were resuspended in one-half of the original volume in a 1:5 to 1:6 dilution of Table 1 lowtox rabbit complement (Cedariane, Hicksville, NY) for an additional 60 min at 37Â°.Cells were washed 3 times in HBSS and resuspended in Specificity of LAK-mediated lysis Effector with following % of lysis" CM for further study. LAK cells or in vitro-sensitized cells were passed over a Lympholyte-M (LSM; Lederle Laboratories, Ontario, Canada) cul- gradient to remove dead cells and debris prior to depletion. All other Fresh C57BL/6 tured in me- cul steps were performed as described above. TargetMCA-102 splenocytesdium1Â±1* tured inIL-253 Chromium Release and Cold Target-Cell Inhibition Assays. A 4-hr 2Â±2 Â±2 chromium release assay was performed as described previously (30). MCA-106 3Â±1 4Â±1 53 Â±1 -20 Â±2 -18 Â±1 For the cold target-cell inhibition assay, nonradioactive (cold) blocking EL-4 43Â±2 C57BL/6 lymphocyte -12 Â±3 -20 Â±5 8Â±2 cells were tested for their ability to inhibit the lysis of MCA-102 tumor C57BL/6 Con A blast 3Â±2 -10 Â±1 58 Â±5 targets by cytotoxic T-cells; 50 Â¿tlof the cold blockers at the desired YACC57BL/61a 20 Â±1 8 Â±1C57BL/672 + concentration were incubated with 50 n\ of effector cells in wells of .bEffectortarget ratio, 100:1 microtiter plate (Catalogue No. 76-311-05; Linbro Chemical Co., Ham- Moon -i- C C

MAY 1984 1947

IL-2 Tablez Effect of serum on lysis by LAK % of lysis versus MCA-102

In fetal calf serum8 In mouse serum 80 130:110:11:1ExperimentExperiment Â±2.9C24.7 Â±3.012.9 70 Â±2.04.3 Â±1.51.5 Â±1.629.3 Â±1.029.9

60 250:110:11:132.4 Â±2.17.1 50 Â±1.22.6 Â±2.3 a IL-2 produced with fetal calf serum 40 6 IL-2 produced with mouse serum. l Â°MeanÂ±S.E. â€¢30

i Table 3 20 Effect of Jurkat IL-2 on LAKactivation of lysis versus MCA-102"45.2 10 J-IL-2C2,500)J-IL-2 (1:1 Â±6.3"14.4 Â±1.84.7 0; Â±2.347.4

-10 (1:25,000)IL-2 Â±1.414.9 +0.21.1 Â±0.938.7

01234 7 (50%)Effectortarget100:110:11:1100:110:11:1100:110:11:1%Â±1.78.9 DAYS Â±0.7-0.8 Chart 1. Cytotoxic activity of LAK cells after 1- to 7-day culture with IL-2 or Â±1.0 CM. Cytotoxicity was determined in a 4-hr chromium release assay at an effec- " 4-hr chromium release assay versus 5 x 103 MCA-102. LAK were activated tortarget ratio of 100:1. Lytic activity on Day 0 refers to that level of lysis obtained for 96 hr prior to assay. by fresh C57BL/6 splenocyte effector cells. " Mean Â±S.E. c J-IL-2, Jurkat IL-2. lymphokines present in IL-2 preparations. In 2 separate experi ments, C57BL/6 lymphocytes cultured in IL-2 produced in 1% Con A blasts, normal syngeneic splenocytes were poorly lysed mouse serum, and that never were in contact with fetal calf by C57BL/6 cells grown in IL-2. To determine the effects of serum, generated LAK cells capable of lysing the MCA-102 tumor differences in cell lysis that were due to intrinsic differences in in a 4-hr chromium release assay (Table 2). Furthermore, highly lysability, DBA/2 splenocytes that were sensitized in vitro to purified, homogeneous Jurkat IL-2, free of other lymphokine C57BL/6 cells were used as effectors against several C57BL/6 activity, was also capable of activating LAK cells at a 1:12,500 target cells. Both MCA-102 and Con A blast cells were more or 1:25,000 dilution (Table 3). Similar levels of lysis were achieved sensitive to lysis by the in vitro allosensitized cells than were when C57BL/6 lymphocytes were activated in highly purified IL- fresh splenocytes (data not shown). 2 produced by the EL-4 high producer tumor cell line (provided Since C57BL/6 splenocytes were not as sensitive to lysis as by Dr. Alfred Chang; data not shown). the tumor target cells, cold-target inhibition studies were per Furthermore, lysis of syngeneic tumor was not effected by formed to determine whether these cells shared a common extensive washing of effector cells after culture in IL-2 or by specificity that could not be detected using lysis in the 51Cr conducting the lytic assay in Â«-methylmannoside (20 ng\/m\; data release assay. Nonlabeled splenic C57BL/6 or unlabeled Con A not shown). blast cells were mixed with the effector cells and assayed at an Thus, it appears that the lytic activity of splenocytes that effectontarget cell ratio of 10:1 on 51Cr-labeled MCA-102 target results from activation in IL-2 is not dependent on exposure to cells using effector cells grown in IL-2. As shown in Chart 3, either Con A, xenogeneic serum, or lymphokines other than IL- both syngeneic normal splenocytes and syngeneic Con-A blasts 2. resulted in no inhibition of the lysis of syngeneic MCA-102 tumor Lymphocytes from Multiple Strains of Mice and Capability even at high blockereffector cell ratios. Cold MCA-102 cells, to Generate LAK. In order to determine if the activation of LAK however, caused strong inhibition of lysis which was directly cells was a strain-specific phenomenon, lymphocytes from mul proportional to the blockereffector cell ratios. Similar results tiple mouse strains were cultured in IL-2 or CM for 5 days and were obtained in 5 experiments. These data suggested that the tested for lytic activity versus both the fresh MCA-102 and the cells activated in IL-2 recognized a determinant on the MCA-102 EL-4 tumor target (Chart 2). Splenocytes from 6 of 6 strains cell surface with either limited or no expression on the normal tested were capable of generating substantial lysis for both the splenocyte cell surface and distinct from the determinant that MCA-102 and EL-4 tumor targets. Cells cultured in CM exhibited resulted in lysis of C57BL/6 Con A blast cells. minimal or no lysis for the EL-4 and MCA-102 target cells. Cold target inhibition studies of the lysis of C57BL/6 Con A Cold-Target and Blocking Studies. In contrast to the signifi blast cells revealed that both fresh 102-tumor cells and Con A cant levels of lysis of the syngeneic 102 tumor and C57BL/6 blast cells were capable of completely inhibiting lysis (Chart 4).

1948 CANCER RESEARCH VOL. 44

CM EFFECTORSIL-2 90 100 oâ€”Â» AKR oâ€”o 80 â€¢â€”â€¢FreshMCA-102 70 Oâ€”o C57 BL/6 splenocytes BALB/c Dâ€”D C57 BL/6 Con-A blasts 60 80 C3H GO- oâ€”o C57 BI/6 o-o 40 60 DBA/2 30 20 ^~c B10-D2 ?-? 10 40

20 0.1 1 10 100 EFFECTORS/TARGETS

O Noneâ€¢â€” MCA-10270 â€¢Fresh o C57 BL/6 splenocytes D-Dblastsv C57 BL/6 Con-A -20 60GO 10 30 100 ÃŽCENTLYSIS8A EFFECTOR:TARGET 403020100-10Oâ€” p,*Jâ€¢cqj"^--o~--Q^*~~^* -v Q Ã§

-iE/T-10:1 â€”

0.05:1 0.5:1 1:1 2:1 4:1 8:1 BLOCKERS:EFFECTORS Charts. Top, lysis of MCA-102 tumor, and C57BL/6 lymphocytes and Con A blasts by splenocytes grown in IL-2. Bottom, blocking of cytotoxicity versus MCA- C/5 102 tumor cells of splenocytes grown in IL-2, as tested in the cold-target inhibition assay (effector:target-cell ratio, 10:1 ). Points, mean of 3 replicates. Bars, S.E. MCA- 102 tumor but not C57BL/6 splenocytes or Con A blasts could inhibit lysis.

Targets: â€¢â€”â€¢FreshMCA-102 D-Q CS7 BL/6 Con-A blasts

-20 10 30 EFFECTOR: TARGET

Chart 2. Lytic activity of splenocytes from 6 strains of mice activated in IL-2 or 10 20 CM for 5 days. The cytotoxicity assay was performed at various effectortarget cell ratios for 4 hr versus the EL-4 (fop) and MCA-102 (bottom) tumor targets. Blockers: A None Con-A blast Targets â€¢â€”â€¢FreshMCA-102 Thus, although both syngeneic MCA-102 and Con A blasts are Oâ€”O C57 BL/6 Con-A blasts lysed by cells grown in IL-2, different determinants are recog nized on these 2 cells. The tumor cell contains the determinant recognized on the blast cells but contains, as well, a different determinant not present on the blast cells. The Precursor Cells of LAK are Nonadherent and Thy-1.2- Positive. Normal C57BL/6 lymphocytes received no treatment or were depleted of adherent cells by incubation on plastic Retri 60r MCA-102 Targets dishes and passage over nylon wool. Comparable levels of lytic 60 activity were generated when intact or adherence-depleted cells 40 30 were cultured in IL-2 for 5 days and tested against the syngeneic 20 MCA-102 or EL-4 tumor target (Table 4), however, no lysis was 10- detected against fresh C57BL/6 lymphocytes. Intact or adher 0 ence-depleted cells demonstrated no detectable level of lytic activity when cultured in CM for 5 days. 0.05:1 0.5:1 1:1 2:1 4:1 8:1 Normal C57BL/6 splenocytes received no treatment, treat BLOCKERS/EFFECTORS Chart 4. Top, lysis of MCA-102 tumor and C57BL/6 Con A blasts by syngeneic ment with rabbit complement, or treatment with a monoclonal splenocytes grown in IL-2. Middle, cold target inhibition of lysis of C57BL/6 Con A anti-Thy-1.2 antibody and complement. In 2 of 2 experiments, blasts by syngeneic splenocytes grown in IL-2 (effectortarget ratio 10:1). MCA- comparable levels of lysis were obtained when responder 102 tumor and Con A blast cells both inhibited lysis of blast targets. Bottom, cold- target inhibition of lysis of MCA-102 tumor by syngeneic splenocytes grown in IL- C57BL/6 lymphocytes were treated with complement or not 2 (effectortarget ratio, 10:1); MCA-102 tumor but not C57BL/6 Con-A blast cells (Table 5). The ability to generate allospecific cytotoxicity from inhibited lysis of 102-tumor.

MAY 1984 1949

Table 4 LAK-activated effector cells were depleted of Thy-1.2+, Lyt-1+, Effect of adherent cell depletion on LAK activation and Lyt-2+ T-cell subsets using monoclonal antibody and com Target of % lysis* plement. Concurrently, a C57BL/6 anti-DBA/2 effector cell pop lympho- ulation was similarly depleted with the identical reagents. After EL-4cyte-2 depletion with monoclonal antibody and complement, viable cells Whole-cell population were counted, and cytotoxicity was determined at known ratios. Incubated in CM + 0 + 3 In 2 independent experiments (Table 6), anti-Lyt-2.2 or anti-Thy- Incubated in IL-2MCA-1023Â±0" 43 Â±3C57BL/6 53 Â±3-5 6Â±4 1.2 monoclonal antibody and complement treatment eliminated Adherence-depleted cells most allogeneic killing of the relevant P-815 target as well as Incubated in CM -4Â±0 -2Â±0 -10 Â±2 LAK activity against the syngeneic 102-tumor target. Anti-Lyt- Incubated in IL-2 35 Â±2 55 Â±2 -7 Â±3 * Effectortarget ratio, 100:1. 1.2 and complement only slightly lowered both activities, while 6 Mean Â±S.E. complement alone had no effect compared to nontreated control cells. These results indicated, therefore, that, functionally, the LAK effector cell is Thy-1.2+, Lyt-1.2~, and Lyt-2.2+. Table 5 Phenotype of LAK precursor cells Effect of Anti-Lyt-2.2 and Anti-LFA-1 on LAK Effector Cell % of lysis Killing. To determine what role Lyt-2 and LFA-1 membrane Allogeneic" LAK" Effectortarget molecules played in LAK cell lysis, LAK cells or allogeneic CTL Experiment1None100:110:11:1Complement100:110:11:1Anti-Thy-1 were pretreated with monoclonal antibody directed against either the Lyt-2 or LFA-1 membrane molecule prior to being used as Â±3.1e20.9 Â±4.320.1 effector cells in a chromium release assay (Table 7). BALB/c Â±1.53.0 Â±3.94.0 anti-C57BL/6 and C57BL/6 anti-BALB/c in w'fro-sensitized effec Â±1.756.7 Â±0.953.8 tor cell populations generated 23 and 75% lysis for the respective Â±3.528.2 Â±5.618.3 EL-4 and P-815 allogeneic tumor targets at a 20:1 effectortarget Â±1.34.9 Â±2.01.9 Â±0.21 +1.02.4 ratio in a 5-hr chromium release assay. Addition of anti-Lyt-2 or complement100:110:11:1Experiment.2 + anti-LFA-1 monoclonal antibody to anti-EL4 allogeneic CTL to .8 Â±0.20.1 Â±0.40.2 tally inhibited lysis of EL-4 at concentrations as low as 1:200 Â±0.70.5 Â±0.5-1.2 and 1:100,000, respectively. Anti H-2" (anti-BALB/c) allogeneic Â±0.685.5 Â±0.223.7 effector cells demonstrated a similar inhibition of lytic activity 2None50:15:10.5:1Complement50:15:10.5:1Anti-Thy-1 when treated in the same manner. LAK cells generated 49% Â±1.8"25.4 Â±4.111.9 lysis for the syngeneic EL-4 target and 42% lysis for the alloge Â±0.51.5 +1.64.1 neic P-815 target at a 20:1 effectortarget ratio. Lytic activity Â±0.684.0 Â±5.629.4 against the syngeneic EL-4 target was inhibited by 57% by the Â±4.7*41 Â±1.59.6 Lyt-2 reagent and completely inhibited by the LFA-1 reagent at .9Â±1.94.0 Â±0.7-5.5 Â±1.02.4 Â±1.15.3 a 1:200 and 1:10,000 dilution, respectively. A similar pattern of complement50:15:10.5:164.0.2 + inhibition was demonstated for the allogeneic P-815 target. Both Â±0.7a1.4 Â±0.42.0 anti-Lyt-2 and LFA-1 had no effect on the spontaneous release Â±0.60.3 Â±0.2-3.1 of the P-815 target, but slightly increased the spontaneous Â±0.450.8 + 2.0 * C57BL/6 splenocytes activated for 96 hr in 50% IL-2 and tested for lysis release of the EL-4 target to -16 and -18%, respectively. Thus, versus 5 x 103 MCA-102 tumor targets. membrane molecules essential to LAK cell lysis reacted with 6 C57BL/6 anti-DBA/2 allogeneic sensitization in complete medium. P-815 tumor anti-LFA-1 antisera and, to a lesser extent, anti-Lyt-2 antisera. was used as the target. c 4-hr chromium release assay. "C57BL/6 anti-DBA/2 allogeneic sensitization in 50% IL-2. " Mean Â±S.E. DISCUSSION The development of techniques for the generation of T-lym- Thy-1.2-depleted cultures was totally eliminated whether the phocytes with cytotoxic reactivity for fresh autologous tumor sensitization was performed in the presence or absence of IL-2, cells has been an elusive goal of many laboratories in recent indicating an effective elimination of Thy-1+ cells. Pretreatment years. Our laboratory has recently reported that the incubation of responding splenocytes with anti-Thy-1.2 and complement of either normal or cancer patients' lymphocytes in IL-2, for as resulted in a total loss of ability to generate LAK activity by short as 48 hr leads to the generation of significant lytic activity exposure to IL-2, indicating that the LAK precursor cell bears for autologous or allogeneic fresh solid tumor (8). Human LAK Thy-1.2. cells are capable of lysing a wide variety of histologically distinct Phenotype of LAK Effector Cell. Phenotypic characterization NK-resistant solid tumor cells, but do not lyse autologous pe by cell sorter analysis (data not shown) indicated that normal ripheral blood lymphocytes, Con A lymphoblasts, or single-cell lymphocytes or lymphocytes infiltrating syngeneic tumor uni suspension from several normal adult tissues (8-10). The pre formly express Thy-1.2 and Lyt-2.2 with low levels of Lyt-1 after cursor of the human LAK cell is distinct from both NK and T- short-term (1 week) culture in IL-2. When bulk cells are main cells based on serological phenotype, density gradient sedimen tained in culture with IL-2 for more than 2 weeks, Lyt-1 is no tation, and tissue distribution, while the human LAK effector cells longer detectable on the cell surface, and all cells under these bears all the phenotypic, morphological, and functional charac culture conditions were Thy-1+, Lyt-2+. In order to determine the teristics of CTL (10). The biological role of these LAK cells in the phenotype of the cell responsible for LAK cell activity, 4-day human is unknown. We have developed an animal model of the

1950 CANCER RESEARCH VOL. 44

TableÃ³ Phenotypeof LAK effector cells % of lysis3

1TreatmentNoneComplementAnti-Lyt-2.2Anti-Lyt-1Experiment 2Allogeneic83.0

Â±6.4" Â±1.7 Â±2.4 Â±2.7 10:1100:1 39.1Â±1.060.0+1.5 8.9 Â±0.738.0 59.3 Â±3.385.2 19.6Â±1.352.2

Â±2.9 Â±3.1 Â±0.3 10:1100:1 23.4 Â±2.42.2 9.2 Â±0.78.6 50.4 Â±4.90.6 19.5Â±0.512.3

Â±0.1 Â±1.5 Â±0.1 Â±0.4 10:1100:1 1.6Â±0.354.9 0.9 Â±0.634.0 2.3 Â±0.576.2 4.2 Â±0.848.3

.2Anti-Thy-1 Â±2.4 Â±2.3 Â±1.2 Â±3.0 10:1100:1 13.5Â±0.61 8.7 Â±0.36.5 56.9 Â±7.30.8 18.5 Â±1.4-0.5

.2Effector:target100:1 .4 Â±0.2 Â±0.6 Â±0.4 Â±0.5 0.3a 10:1Allogeneic076.60.2 Â±0.8LAKC30.70.4 Â±0.5Experiment-1.4 Â±0.3LAK59.3-0.7 Â± 4-hr chromium release assay. 6 5 x 103P-815 (H-2") target. c 5 x 103MCA-102 (H-2') target. d Mean Â±S.E.

Table 7 Effect of anti-Lyt-2 or anti-LFA-7 on LAK or allogeneic-mediatedlysis of inhibÃ P-815% of inhibi lysis3NoneAnti-Lyt-2Anti-LFA-1NoneAnti-Lyt-2Anti-LFA-1NoneAnti-Lyt-2Anti-LFA-1Anti-Lyt-2Anti-LFA-1,20:1.1:2,0001:2001:201:100,0001:10,0001:1,0001:2,0001:2001:201:100,0001:10,0001:1,0001:2,0001:2001:201:100,0001:10,0001:1,0001:2001:1EL-4 % of EffectorC57BL/6 tion0-17130187161196213051577665145127-oflysis874 tion0414461965760193350127483 anti-BALB/cBALB/c Â±3*71 Â±564Â±040

Â±460 Â±326 Â±118Â±142

anti-C57BL/6C57BL/6 Â±27 Â±-7Â±-20 Â±-14 Â±-22 Â±-26 Â±49

LAK* Â±24 +134 Â±21 Â±028 Â±12Â±17 Â±121 Â±437 Â±-22 Â±011 Â±-13 Â±37Â±00-1% Â±-16

Â±-18 0-423 Â±618721253.5561172831% ratio0Effectortarget Moor. -1- C CTreatment

LAK phenomenon to aid in studies of the role of LAK cells in was not dependent on Con A or lymphokine other than IL-2 vivo. present in crude supernatants, since lymphocytes cultured in In this report, we have demonstrated that normal nonsensi- highly purified human- or mouse-derived 11-2were also lytic for tized murine splenocytes cultured in IL-2 for as short as 2 days tumor. Furthermore, we3 and others (1) have been unable to developed lytic activity for fresh NK-resistant syngeneic tumor induce lytic activity by exposing C57BL/6 lymphocytes to Con A and autologous Con A blasts, but not normal lymphocytes. This under a wide variety of culture conditions in the absence of reactivity was maximal after 4 days of culture of lymphocytes in exogenous IL-2. No differences in the lytic activity of LAK cells IL-2. LAK cells displayed broad lytic reactivity for syngeneic and were observed when the cytotoxicity assay was performed in a- allogeneic tumor or Con A blast targets. Furthermore, activation methylmannoside, indicating no dependence on lectin in the of LAK cells appears to be a general phenomenon in murine target cell recognition by LAK cells. strains, since all strains tested (Chart 2) were capable of gener Several investigators have reported the nonspecific activation ating lysis for LAK-sensitive, NK-resistant tumor targets upon IL- of T-lymphoid cells by fetal calf serum determinants (16, 21, 36). 2 stimulation. The activation of LAK cells to cause lysis of syngeneic tumor 3A. Mazumder, M. Rosenstein,and S. A. Rosenberg,submitted for publication.

MAY 1984 1951

Although fetal calf serum is often present in our IL-2 preparation, assay. This phenomenon may be mediated through a common LAK activation is not dependent on the presence of xenogeneic IL-2 activation pathway, since Lyt-1+ cells are required to gen serum. Lymphoid cells cultured in IL-2 produced in 1% mouse erate this highly lytic cell type. serum exhibited lysis similar to that of LAK activated in fetal calf Other phenomena for the activation of cells lytic for tumor serum containing IL-2 preparations. In each of our experiments, have been described such as natural killing (11,14, 26), natural control cultures of splenocytes incubated in CM containing fetal cytotoxicity (23, 33), activated cell killing (18, 19), activation by calf serum were included, and in no instance was significant lytic serum determinants (16, 21, 36, 39), 2-mercaptoethanol (12), activity generated. tumor necrosis serum (2), polyinosinic acid (3), allogeneic stim We have used monoclonal antibody and complement depletion ulation, or specific mixed-lymphocyte tumor cell reactions (4, 5, techniques to characterize the cell surface phenotype of cells 20, 22). In most of these studies, lytic activity was tested against activated in IL-2. The LAK effector cell is sensitive to treatment cultured instead of the fresh tumor target cells used in our with anti-Thy-1.2 or anti-Lyt-2.2 monoclonal antibody and com studies. plement, but only slightly sensitive to anti-Lyt-1.2 antibody and Of interest is the finding that T-cells activated in culture by IL- complement. NK cells are thought to express Lyt-1 and low 2 do not lyse fresh lymphoid cells. Since fresh lymphocytes are levels of Thy-1 (6,13), whereas conventional CTLs express high less lysable than tumor cells, we have attempted to use cold- levels of Thy-1 and Lyt-2 but not Lyt-1 on their cell surfaces (34, target inhibition studies to determine if fresh lymphocytes share 35). As demonstrated above, the LAK cell expressed a conven recognition determinants with tumor cells that are lysed by LAK. tional CTL phenotype (Thy-1+ Lyt-1 '2*) and, therefore, appears Although tumor cells, by nature of their larger size, are more distinct from NK cells. Another important difference between efficient cold-target inibitors, we have never seen inhibition of LAK cells and NK cells is the ability of LAK cells to kill a variety MCA-102 killing at even very high numbers of unlabeled fresh of fresh tumor target cells that are resistant to NK cell lysis. lymphocytes or Con A blast cells, suggesting, therefore, that a These findings are similar to those we have reported in the component on the MCA-102 cell surface that is recognized by human in which the LAK cell is sensitive to OKT-3 and OKT-8, LAK is absent or poorly expressed on the normal splenocyte cell but not to OKT-4 antibodies (8), and also lyses NK-resistant surface. The surprising observation that Con A blast targets are fresh human tumor cells. lysed by LAK, but do not block LAK killing of MCA-102 at high Further evidence that the murine LAK cell expresses the cold-target inhibitor ratios, possibly reflects the heterogeneous characteristics of CTL and not NK cells was the finding that both nature of the LAK cell population. Cloning of the LAK cell Lyt-2 and LFA-1 membrane antigens are involved in the cytolytic population will further unravel this phenomenon. activity of LAK effector cells. Treatment of LAK cells or allogeneic The lysis of fresh tumor cells by syngeneic lymphoid cells CTL with anti-Lyt-2 or LFA-1 monoclonal antibody during cyto- incubated in IL-2 was an unexpected finding and is not well toxicity assays inhibited killing of the relevant syngeneic or understood. The high lysis of syngeneic solid tumor cells that allogeneic tumor target in a dose-dependent fashion. Reports by we have seen using a short-term cytotoxicity assay raises the other laboratories (7) demonstrated no effect of anti-Lyt-2 on the possibility that these cells may be useful in in vivo studies of blocking of NK activity, although concentrations of anti-LFA-1 in immune lysis by cytotoxic lymphoid cells. We have previously excess of those used in this report inhibited NK lytic activity by demonstrated that human cells grown in IL-2 exhibited lysis of approximately 50%. autologous tumor and that mouse and human cells grown in IL- The precursor of the LAK cell appears to be a nonadherent 2 can be safely reinfused into syngeneic or autologous hosts. Thy-1+ cell. This result is distinct from LAK generated in the The potential immunotherapeutic effect of cells expanded in IL- human system (8-10), where no marker for the LAK precursor 2 in mouse tumor models is currently under investigation. has yet been detected. While the lineage of the LAK cell in murine and human systems may be different, it is more likely that human REFERENCES and anti-T-cell reagents recognize determinants with different characteristics than those recognized by anti-Thy-1 antibodies 1. Bonavida, B., and Bradley, T. P. Studies on the induction and expression of T cell mediated immunity. V. Lectin-induced nonspecific cell mediated cytotox in the mouse. Preliminary experiments with anti-Lyt-1.2 or Lyt- icity by alloimmune lymphocytes. Transplantation (Baltimore), 27: 94-102, 2.2 inhibits the generation of LAK. Experiments are currently in 1976. 2. Chun, M., Pasanen, V., Hammerting, U., Hammerling, G. F., and Hoffman, M. progress to determine if this is due to the elimination of a single K. Tumor necrosis serum induces a serologically distinct population of NK precursor subset or if LAK activation requires the interaction of cells. J. Exp. Med., 150: 462-431, 1976. various subpopulations of lymphoid cells. 3. Dorfman, N., Winkler, D., Burton, R. C., Kossayda, N., Sabia, P., and Wun derlich, J. Broadly reactive murine cytotoxic cells induced in vitro under Activated lymphoid cells appear to acquire the capacity to lyse syngeneic conditions. J. Immunol., 729:1762-1769, 1982. a wide variety of target cells depending on the activating stimu 4. Eberlein, T. J., Rosenstein, M., and Rosenberg, S. A. Regression of a dissem inated syngeneic solid tumor by systemic transfer of lymphoid cells expanded lus. Paciucci ef a/. (22) and Zarling ef a/. (38) have reported that in interleukin 2. J. Exp. Med., 756: 385-397,1982. murine and human cells activated by alloantigens are capable of 5. Eberlein, T. J., Rosenstein, M., Spiess, P. J., Wesley, R., and Rosenberg, S. killing autologous tumor cells. Bonavida and Bradley (1), how A. Adoptive chemoimmunotherapy of a syngeneic murine lymphoma using long temi lymphoid cell lines expanded in T cell growth factor. Cancer Immunol. ever, claimed that murine cells activated by alloantigens can lyse Immunother, 13: 5-13, 1982. syngeneic tumor only in the presence of mitogen and that these 6. Glimcher, L, Shen, F. U., and Cantor, H. Identification of a cell-surface antigen expressed on natural killer cells. J. Exp. Med., 745: 1-9, 1977. cells appear to be identical to the cells lytic for alloantigens. 7. Golstein, P., Pierres, M., Schmitt-Verhulst, A., Luciani, M., Bufeme, M., Eshhar, Mazumder ef a/.3 have recently shown that splenocytes from Z., and Kaufmann, Y. Functional relationships of lymphocyte membrane struc selected mouse strains can lyse fresh autologous NK-resistant tures probed with cytolysis and/or proliferation inhibiting H35-27.9 and H35- 89.9 monoclonal antibodies. In: P. Golstein and W. R. Clark, (eds.), Mecha tumor when incubated for 3 days with Con A, and that their lytic nisms of Cell-Mediated Cytotoxicity, pp. 481-502. New York: Plenum Publish activity did not require the presence of Con A in the cytotoxicity ing Corp., 1982.

1952 CANCER RESEARCH VOL. 44

8. Grimm, E. A., Mazumder, A., Zhang, H. Z., and Rosenberg, S. A. The 24. Parker, G. A., and Rosenberg, S. A. Serologie identification of multiple tumor- lymphokine activated killer cell phenomenon: lysis of NK-resistant fresh solid associated antigens on murine sarcoma. J. Nati. Cancer Inst., 58:1303-1309, tumor cells by IL-2 activated autologous human peripheral blood lymphocytes. 1977. J. Exp. Med., 755:1823-1841,1982. 25. Pierres, M., Gordidis, C., and Goldstein, P. Inhibition of murine T cell-mediated 9. Grimm, E. A., Ramsey, K. M., Mazumder, A., Wilson, D. J., Djeu. J. Y., and cytolysis and T cell proliferation by a rat monoclonal antibody immunoprecipi- Rosenberg, S. A. Lymphokine-activated killer cell phenomenon. II. Precursor tating two lymphoid cell surface polypeptides of 94,000 and 180,000 molecular phenotype is serologically distinct from peripheral T lymphocytes, memory weight. Eur. J. Immunol., 72: 60-69,1982. cytotoxic thymus-derived lymphocytes and natural killer cells. J. Exp. Med., 26. Roder, J. C., Karre, K., and Kiessling, R. Natural killer cells. Prog. Allergy, 28: 757. 884-897, 1983. 66-159,1981. 10. Grimm, E. A., and Rosenberg, S. A. The human lymphokine-activated killer 27. Rosenberg, S. A., Schwarz, S., and Spiess, P. J. In vitro growth of murine T cell phenomenon. In: E. Pick, and M. Candy (eds.), Lymphokines, Vol. 9. New cells. II. Growth of in vitro sensitized cells cytotoxic for alloantigens. J. York: Academic Press, 1983. Immunol., 727:1951-1955, 1978. 11. Herberman, R. B. Natural killer cells and cells mediating antibody dependent 28. Rosenberg, S. A., Spiess, P. J., and Schwarz, S. In vitro growth of murine T cellular cytotoxicity against tumors. Clin. Immunobiol., 4: 73,1980. cells. I. Production of factors necessary for T cell growth. J. Immunol., 727: 12. Igarashi, R., Okada, M., Kishimoto, T., and Yamamura, Y. In vitro induction of 1946-1950, 1978. polyclonal killer T cells with 2-mercaptoethanol and the essential role of 29. Rosenstein, M., Ebenem, T., Kemeny, M. M., Sugarbaker, P. H., and Rosen macrophages in this process. J. Immunol., 778:1697-1703,1977. berg, S. A. In vitro growth of murine T cells. VI. Accelerated skin graft rejection 13. Kiessling, R., Klein, E., and Wigzell, H., "Natural" killer cells in the mouse. I. caused by adoptively transferred cells expanded in T cell growth factor. J. Cytotoxic cells with specificity and distribution according to genotype. Eur. J. Immunol., 727: 566-571, 1981. Immunol., 5: 112-117,1975. 30. Rosenstein, M., Eberlein, T., Schwarz, S., and Rosenberg, S. A. Simplified 14. Kiessling, R., and Wigzell, H. An analysis of the murine NK cell as to structure, techniques for the isolation of alloreactive cell lines and clones with specific function and biological relevance. Immunol. Rev., 44:165-208,1979. cytotoxic or proliferative activity. J. Immunol. Methods, 67: 183-193,1983. 15. Ledbetter, J. A, and Herzenberg, L. A. Xenogeneic monoclonal antibodies to 31. Spiess, P. J., and Rosenberg, S. A. A simplified method for the production of mouse lymphoid differentiation antigens. Immunol. Rev., 47: 63-90,1979. murine T-cell growth factor free of lectin. J. Immunol. Methods, 42: 213-222, 16. Levy, R. B., Shearer, G. M., Kim, K. J., and Asofsky, R. M. Xenogeneic serum- 1981. induced murine cytotoxic cells. I. The generation of effector components 32. Strausser, J. L., and Rosenberg, S. A. I. Growth of cells sensitized in vitro to specific for self and allogeneic target cells. Cell. Immunol., 48:276-287,1979. alloantigens. J. Immunol., 727:1491-1495, 1978. 17. Lotze, M. T., Grimm, E. A., Mazumder, A., Strausser, J. L., and Rosenberg, 33. Stutman, O., Paige, C. J., and Figarella, E. F. Natural cytotoxic cells against S. A. Lysis of fresh and cultured autologous tumor by human lymphocytes solid tumors in mice. I. Strain and age distribution and targets cell susceptibility. cultured in T-cell growth factor. Cancer Res., 41: 4420-4425,1981. J. Immunol., 727: 1819-1826, 1978. 18. Masucci, M. G., Klein, E., and Argov, S. Disappearance of the NK effect after 34. Stutman, O., and Shen, F. W. H-2 restriction and non-restriction of T-cell- explanation of lymphocytes and generation of similar nonspecific cytotoxicity mediated cytotoxicity against mouse mammary tumour targets. Nature (Lond.), con-elated to the level of blastogenesis in activated cultures. J. Immunol., 724: 276:181-182,1978. 2458-2463,1980. 35. Stutman, 0., Shen, F. W., and Boyse, E. A. Ly phenotype of T cells cytotoxic 19. Mazumder, A., Grimm, E. A., and Rosenberg, S. A. Characterization of the for syngeneic mouse mammary tumors: evidence for T cell interactions. Proc. lysis of fresh human solid tumors by autologous lymphocytes activated in vitro Nati. Acad. Sci. U. S. A., 74: 566, 1977. with phytohemagglutinin. J. Immunol., 730: 958-964,1983. 36. Thorn, R. M. Murine T-cell mediated cytotoxicity against syngeneic and allo 20. Minato, N., Reid, L., and Bloom, B. R. On the heterogeneity of murine natural geneic cell lines induced by fetal calf serum. Cell. Immunol., 54: 203-214, killer cells. J. Exp. Med., 754: 750-762,1981. 1980. 21. Ortaldo, J. R., Bonnard, G. D., and Herberman, R. B. Cytotoxic reactivity of 37. Yron, J., Wood, T. A., Spiess, P. J., and Rosenberg, S. A. In vitro growth of human lymphocytes cultured in vitro. J. Immunol., 779: 1351-1357, 1977. murine T cells. V. The isolation and growth of lymphoid cells infiltrating 22. Paciucci, P. A., Macphail, S., Zariing, J. M., and Bach, F. H. Lysis of syngeneic syngeneic solid tumors. J. Immunol., 725: 238-245,1980. solid tumor cells by alloantigen stimulated mouse T and non-T cells. J. 38. Zariing, J. M., and Bach, F. H. Continuous culture of T cells cytotoxic for Immunol., 724: 370-375, 1980. autologous human leukemia cells. Nature (Lond.), 280: 685-688, 1979. 23. Paige, C. F., Figarella, E. F., Cuttitto, M. J., Cahan, A., and Stutman, O. Natural 39. Zielske, J. V., and Golub, S. H. Fetal calf serum-induced blastogenic and cytotoxic cells against solid tumors in mice. II. Some characteristics of the cytotoxic responses of human lymphocytes. Cancer Res., 36: 3842-3846, effector cells. J. Immunol., 727: 1827-1835, 1978. 1976.

MAY 1984 1953

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1984 American Association for Cancer Research. Lymphokine-activated Killer Cells: Lysis of Fresh Syngeneic Natural Killer-resistant Murine Tumor Cells by Lymphocytes Cultured in Interleukin 2

Maury Rosenstein, Ilana Yron, Yael Kaufmann, et al.

Cancer Res 1984;44:1946-1953.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/44/5/1946

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/44/5/1946. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.