Lymphokine-Activated Killer Activity in Long-Term Cultures with Anti-CD3 Plus Interleukin 2: Identification and Isolation of Effector Subsets1

(CANCER RESEARCH 49. 963-968. February 15. 1989] Lymphokine-activated Killer Activity in Long-Term Cultures with Anti-CD3 plus Interleukin 2: Identification and Isolation of Effector Subsets1

Augusto C. Ochoa,2 Diane E. Hasz, Rebecca Rezonzew, Peter M. Anderson, and Fritz H. Bach

Immunobiology Research Center, University of Minnesota Hospital and Clinic, Minneapolis, Minnesota 55455

ABSTRACT short-term LAK cultures, while CD3+ cells from such cultures have low lytic activity against NK-resistant targets (8-11). Peripheral blood lymphocytes cultured in recombinant interleukin 2 We have recently reported the generation of large numbers during 3 to 5 days (short-term cultures) develop the ability to lyse natural of cells with LAK activity using long-term (14 to 21 day) killer-resistant tumor lines and fresh tumor cells, i.e., express lympho- cultures of cells stimulated with anti-CD3 (OKT3) moAb and kine-activated killer (LAR) function. Phenotypic analysis has shown these cells to be natural killer cells, i.e., CD16+ and/or Leu 19+ cells. IL2 (6). Cells stimulated with OKT3 + IL2 undergo a high CD3*,CD16~ T-cells, instead, develop very low LAK function in these increase in cell number while maintaining specific LAK activity cultures. comparable to that of cells cultured for 3 to 5 days. LAK We recently reported the development of long-term (up to 21 days) activity of cells stimulated with OKT3 + IL2 for 14 days is cultured cells with LAK activity by stimulation with OKT3 + interleukin significantly increased when incubated for the last 48 h of 2 (11.2).These culture conditions repeatedly resulted in a several hundred culture in medium containing ILl-ÃŸ,IFN-7, or -ÃŸinaddition fold expansion in cell number. Specific LAK activity on Day 14 of culture to IL2. was comparable to that of 3-day LAK cultures and could be further Results of isolation and identification of the effectors medi enhanced by the addition of Â¡nterleukin1/3, ÃŸ-,or -y-interferon. Total ating LAK activity in long-term cultures with OKT3 + IL2 LAK activity was greatly increased in OKT3 + IL2 cultures over that found in short-term cultures. are presented in this report. In contrast to LAK effector cells Isolation of effectors mediating LAK function in long-term cultures in short-term cultures (3 to 5 days), which predominantly stimulated with OKT3 + IL2 showed that both CD3+,CD16" cells and are CD3~ cells, i.e., CD16+,Leul9+, effectors in long-term cul CD16*,CD3~ cells tested on Day 14 of culture expressed equivalent levels tures include at least two additional subpopulations, of LAK activity as shown by lysis of natural killer-resistant targets, CD3+,CD4-,CD8- cells and Leu 19+,CD3~,CD16- cells. The III 60 and Daudi. Further dissection of the subpopulations developing two major T-cell subpopulations, CD3*,CD4+ and CD3+,CD8+ LAK activity demonstrated that, in addition to CD16*,CD3~ cells, CD3+, lymphocytes, develop significantly lower LAK activity. CD4-,CD8- cells and Leu 19+,CD3-,CD16- cells also developed high LAK activity in long-term cultures with OKT3 + IL2. Further, long-term culture with OKT3 + IL2 induced increases in the numbers not only MATERIALS AND METHODS of CD3*,CD4-,CD8- cells but also of CD16%CD3" and Leu 19+,CD3~,CD16~ cells. Although there is a significant increase in the Isolation and Culture of LAK Cells. PBL were isolated from heparin- number of CD3*,CD8+ cells, neither these, nor the CD3+,CD4+ cells, ized venous blood by centrifugaron over Ficoll-Hypaque. Isolated mediate LAK activity to the same extent as the populations mentioned mononuclear cells were washed 3 times with PBS (GIBCO Laborato above. ries, Grand Island, NY) and counted. Cells (1 x IO6)were cultured in 16-mm wells (Costar No. 3424; Costar, Cambridge, MA) in 2 ml of TCM consisting of RPMI 1640 supplemented with 25 mM 4-(2-hy- INTRODUCTION droxyethyl)-l-piperazineethanesulfonic acid, 2 mM L-glutamine, 100 PBL3 stimulated by alloantigens (1, 2) or cultured in either units per ml of penicillin. 100 ng/ml of streptomycin (GIBCO, Grand Island. NY), and 6% pooled heat-inactivated human serum. Highly interleukin 2-containing supernatants or medium with IL2 for purified IL2 from Escherichia coli (12, 13) (Cetus Corporation, Em 3 to 5 days (3-5) develop lytic activity against fresh tumor cells eryville, CA) was used at 1000 units/ml. Cultures were incubated at and several NK-resistant targets, such as Daudi and HL60 (6). 37Â°Cin a humidified atmosphere of 5% CÃ›2.Cell density was deter These cells have been named LAK cells. Initial reports sug mined by counting cells every 48 h. Cells were subcultured in fresh gested that LAK precursor cells did not express the T-cell TCM plus IL2 at 0.5 x IO6cells/ml. Cultures which were treated with receptor as determined by anti-CD3 binding, but that after anti-CD3 moAb (OKT3; Ortho, Raritan, NJ) were initiated as above, i.e., 0.5 x IO6 PBL/ml plus 1000 units/ml of IL2; however, 10 ng/ml incubation in IL2, the effector population expressed CD3 (7). of anti-CD3 were present in culture during the first 48 h. At this time More recent reports have clearly demonstrated that cells ex the cells were counted and subcultured at 0.1 to 0.2 x IO6cells/ml in pressing NK markers, such as CD 16 (Leu 11) and/or Leu 19, TCM containing IL2; i.e., the anti-CD3 moAb was added only during are responsible for the great majority of the LAK activity of the initial 48 h of culture and not thereafter. Tumor Lines. Tumor lines K562, HL60, and Daudi were maintained Received 4/1/88; revised 8/1/88. 11/7/88; accepted 11/14/88. The costs of publication of this article were defrayed in part by the payment in culture in RPMI 1640 with 10% fetal bovine serum (GIBCO, Grand of page charges. This article must therefore be hereby marked advertisement in Island, NY). Cells were subcultured at 0.5 x 106/ml in fresh medium accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This is Paper 478 from the Jordan Bazelon Laboratories of the Immuno- twice a week. Cells of line HL60 are not lysed by unstimulated PBL and are therefore considered NK resistant. Although cells of some biology Research Center, Box 724, University of Minnesota, Hospital and Clinic, Harvard Street at East River Road. Minneapolis. MN 55455. This work was donors show the ability to use the Daudi cells to a low extent (generally supported in part by NIH Grants AI 17687. AI 18326. AI 19007. AI 22682. AI <10% lysis), it is usually also considered as NK resistant. 72626. and CA 47097. Cell-mediated Lympholysis. This was done as described elsewhere 2To whom requests for reprints should be addressed at: Immunobiology (14). In summary, tumor cell line targets were labeled with 250 to 750 Research Center. Box 724 UMHC, 420 Delaware St. S.E., Minneapolis. MN tiC'i of Na5'CrO4 (5000 /jCi/ml; New England Nuclear, Boston, MA) 55455. 3The abbreviations used are: PBL, peripheral blood lymphocytes; IL2, inter for 1 to 1'/2 h at 37Â°C.Cells were washed once in TCM, resuspended leukin 2; rIL2. recombinant IL2; moAb. monoclonal antibody: LAK, lymphokine- in culture medium and held at room temperature for 60 min. Cells were activated killer; NK. natural killer; PBS, phosphate-buffered saline: TCM, tissue then washed twice, resuspended in fresh medium, counted, and ali- culture medium: FITC. fluorescein isothiocyanate: PE. phycoerythrin; ILI. inter leukin 1; 1FN. interferon; MHC, major histocompatibility complex; CTL, cyto- quoted at 500 targets/well in a 96-well V-bottomed plates (Costar) into toxic T-lymphocytes. which the effectors had been previously aliquoted at set concentrations. 963

The effectortarget cell ratios ranged from 30:1 to 0.1:1. Plates were 70 Daudi centrifuged at 65 x g for 5 min and incubated in 5% CO: at 37Â°Cfor 60 4 h, after which 150 p\ of medium were harvested from each well into a scintillation vial with 3 ml of scintillation fluid (Biofluor; New 50 England Nuclear, Boston, MA); radioactivity was counted in a liquid 40 scintillation counter (LKB 1216). Cytotoxicity was determined by 30 % of cytotoxicity 20 _ experimental mean cpm-spontaneous release mean cpm 10 x 100 maximal mean cpm-spontaneous mean cpm 0 10:1 3:1 0.3:1

Cell Sorting by Immunofluorescence. PBL were cultured in OKT3 + 70 HL60 IL2 as described elsewhere (6). After 12 days in culture, cells were counted and adjusted to 1 x 107/ml. As many as 60 x IO6cells were 60 incubated with FITC-conjugated or PE-conjugated moAbs [OKT3, 50 OKT8, OKT4 (Ortho Diagnostics, Raritan, NJ) or Leu 2 (CDS), Leu 40 3 (CD4), Leu 4 (CD3), Leu 5b (CD2), Leu 11 (CD 16), Leu 19 (NKH1) (Becton Dickinson, Mt. View, CA)] for 30 min at 4'C. Cells were then o 30 washed twice with PBS containing 2% fetal calf serum. Cells were * 20 sorted on a FACS IV (Becton Dickinson, Mt. View, CA). Sorted cells were centrifuged, and an aliquot was restained to test the purity of the 10 populations. All of the sorted subpopulations used for determining 0 LAK activity were more than 97% positive for the desired cluster 10:1 3:1 1:1 0.3:1 determinant (see "Results"). Some experiments included three-color analysis using biotin-conjugated Leu 4 (CD3). Sorted cells were resus- K562 pended in RPMI + 1L2 + 6% human serum and kept at 37Â°Cuntil 70 r tested for LAK activity. All sorted subpopulations were analyzed by 60 immunofluorescence to test the effectiveness of the separation. 50 Cell Separation by Panning. Isolation of subpopulations not express ing CD 16 required the depletion of CD16*,CD3~ cells from cultured 40 PBL prior to fluorescence sorting. Two cycles of panning were done 30 following the method of Wysocki and Sato (15). Cells were counted and adjusted to 1 x 107/ml, incubated with the appropriate concentra 20 tion of moAb for 45 min at 4'C, and washed 3 times with 2% fetal calf 10 serum-PBS. Polystyrene, bacteriological grade Petri dishes, 15 x 100 mm (No. 8-757-13; Fisher, Eden Prairie, MN), were coated with goat anti-mouse IgG (Tago, Burlingame, CA). Twenty to 30 x 10' antibody- 10:1 3:1 1:1 0.3:1 Effector Target coated cells were distributed in the plates in 5 ml of PBS. Plates were incubated for 70 min at 4'C. Nonadherent populations were then gently Fig. 1. PBL stimulated with OKT3 (10 ng/ml) on Day 0 were cultured for 13 days in II 2 containing medium. Cells were then positively selected by fluores poured off the plate, washed 3 times in PBS, and used for further cence-activated cell sorting using moAbs OKT3 (anti-CD3) and Leu 11 (anti- separation. Adhering cells were not used in the experiments. The purity of the depletion was assessed by fluorescence (see "Results"). Combi CD 16). Cells recovered were cultured overnight in IL2-containing medium and tested for LAK activity 18 h later. nation of panning and sorting was used to obtain the cleanest possible cell preparations and still have sufficient cells to conduct adequate in 4 of 9 experiments (one of which is shown in the figure) and experiments. equal to in 5 of 9. Essentially equivalent LAK activity between CD3+,CD16~ cells and CD16+,CD3" cells was observed in 7 of 9 experiments. In 2 of 9 experiments, CD16+,CD3~ cells had RESULTS higher LAK activity than CD3+,CD16~ cells when tested CD3*,CD16- Cells and CD16+,CD3- Cells Develop Similar against the same targets, although CD3+,CD16" cells were Levels of LAK Activity in Long-Term Culture. PBL stimulated significantly lytic on all targets (data not shown). When tested against K562, CD16+,CD3~ cells always showed higher lytic with OKT3 and cultured in IL2 for 12 days were separated by FACS sorting into CD3+ cells or CD16+ cells. After sorting, activity than CD3+,CD16~ cells. aliquots of cells were restained with moAbs to determine the To further identify and isolate cells developing LAK activity purity of these subpopulations. Cells sorted for CD3 were more in long-term OKT3 + IL2, we depleted CD 16* cells by panning. than 98% CD3+ and less then 1% CD16+; cells sorted for CD16 Contaminating CD16+,CD3~ cells were less than 1% after two were over 94% CD16+ and less than 1% CD3+. Less than 5% cycles of panning from Day 13 PBL cultures which had been of cells were doubly positive for CD3 and CD 16 in any one stimulated with anti-CD3 + IL2. The remaining cells were experiment. labeled with anti-CD4 and anti-CD8 moAbs for sorting. As Following sorting, cells were cultured overnight in IL2-con- seen on Fig. 2, three subpopulations could be established: CD4+ taining medium at 37Â°Cand tested for LAK activity 18 h later. cells; CDS* cells; and CD4~,CD8~ cells; i.e., cells not staining In several experiments, cells were also tested for lytic activity with either of these antibodies. After sorting, the different 48 h after the sort. LAK activity was defined as the ability to subpopulations were immediately tested for LAK and NK ac lyse the NK-insensitive tumor lines HL60 and Daudi. NK tivity, and aliquots of each subpopulation were reanalyzed to activity was defined by lysis of K562. As seen in Fig. 1, both assess the purity of the sorted subpopulations. Results shown CD3+,CD16~ cells and CD16+,CD3" cells had high lytic activ in Fig. 3 established that the majority of LAK activity was present in the CD4",CD8" population, while CD3+,CD8+ and ity against HL60 and Daudi. LAK activity is comparable in both populations and was higher than in the unseparated PBL CD3+,CD4+ cells had much lower LAK function. The latter 964

I,I 63

63 FITC LEU 5B FITC Fig. 2. Day 13 cultured PBL were depleted of CD16 cells by panning and then Fig. 5. The CD3\CD

Table 1 CD3*,CD4~,CD8~ LAK effectors do not bind WT3I and bind Daudi anti-Ã²moAb >of positive cells with the moAb Cell WT3T anti-o0 CD3\CD4-,CD8~ <1 31 " T-cell receptor 1 moAb (Becton Dickinson, Mountain View, CA). 'TCS-1 (T-Cell Sciences, Cambridge, MA).

30:1 10:1 3:1 1:1 .3:1 .1:1 within the cells not staining for CD4 or CDS; 30% of the cells were Leu 19+,CD3~, while 60% of the cells were CD3+,Leu 19" (similar data was obtained in three other experiments). Further analysis showed that approximately 70% of the Leu 19+,CD3~ cells expressed CD2 (Leu 5B) (Fig. 5). These cells did not U express CD 16 (data not shown). Further analysis of the other f subpopulation, i.e., CD3+,CD4~,CD8~ cells, is presented in Table 1. Sorted cells were restained with a moAb directed against the aÃŸT-cell receptor (WT31; Ortho, Raritan, NJ) or against the d chain (TCS-1; T Cell Sciences) of the yÃ´T-cell receptor. None of the CD3+,CD4~,CD8~ cells bound WT31 80 â€¢¿ moAb, while approximately 30% of these cells bound anti-5 60 - moAb. The anti-5 mo Ab used (Â¿TCS1)identifies a subpopula-

40 tion within the 75 cells (T Cell Sciences technical bulletin), therefore explaining the relatively low percentage of positive 20 cells. Fig. 6 shows LAK activity in these two subpopulations. Both 30:1 101 31 11 .3:1 Leu 19+,CD3-,CD16- and CD3+,CD4-,CD8" cells had as high Effector Target Fig. 3. PBL cultured for 13 days were depleted of CD 16 cells and were further specific LAK activity as unseparated PBL. Thus, three dif sorted using PE-Leu 2 (CDS) and FITC-Leu 3 (CD4) moAbs. Cells were tested ferent subpopulations could be clearly identified as having for LAK activity immediately after the sort. Undepleted PBL were used as control. developed LAK activity in long-term cultures with OKT3 + IL2: (a) CD16+,CD3~; (b) Leu 19+,CD3-,CD16~; and (c) CD3+,CD4-,CD8~ cells. 63 Cells Mediating LAK Activity Increase in Number in Long- Term Culture with OKT3 + IL2. PBL cultured in IL2 or IL2 + OKT3 for 14 days were phenotyped for CD3+ or CD16+ cells on Days 0 and 14. Table 2 shows the increase in the number of CD3+ cells and CD16+ as calculated based on the total number of cells and the percentage of cells expressing the CD3 or CD 16 markers on Days 0 and 14. Five million PBL were cultured in IL2 with or without OKT3. Cells cultured in IL2 alone had a total increase of 9.8-fold (from 5 x IO6to 49 x 10") in 14 days. On the first day of culture, 46% (2.3 x IO6)were CD3+ and 8% (0.4 x 10") were CD16+. On Day 14, 62% (30 x 10") of the Fig. 4. Sorted cells not staining for CD4 or CDS (Fig. 2) were restained with several moAbs. Positive staining was observed with PE-Leu 19 and FITC-Leu 4 cells were CD3+ and 14% (7 x 10") were CD16+, resulting in a (CD3) (right). Control cells were stained with PE or FITC goat anti-mouse 13-fold increase of CD3+ cells and a 17-fold increase of CD16+ immunoglobulin (left). cells. When cultured in OKT3 + IL2, this effect was magnified in terms of the increase in cell number. The total cell number two populations were only capable of lysing a LAK target increase for this experiment was 49-fold (from 5 x IO6 to 245 (>20% lysis) in one of five experiments. x IO6); CD3+ cells increased 90-fold, while CD16+ cells in Fig. 4 shows the phenotypic analysis of the populations creased 88-fold. Thus, the use of OKT3 did not exclusively 965

Daudi Phenotypic analysis of other subpopulations developing LAK activity, i.e., CD3+,CD4-,CD8~ and Leu 19+,CD3-,CD16- cells, required 3-color analysis. As shown in Table 3, there was a significant increase in the proportion of these cells over the time of culture. On the first day of culture (Day 0), 7% of the unstimulated cells were CD3+,CD4-,CD8~, and 3% of the Leu 19 cells were Leu 19+,CD3~,CD16~. On Day 14 of culture, the CD3+,CD4~,CD8~ subpopulation had increased to 24%; Leu 19+,CD3~,CD16~ cells increased from 3% to 13%. Further 10:1 3:1 .3:1 analysis of the Leu 19* cells revealed that 76% coexpressed 60 r HL60 50 - " "S CD2. The results of 2 more experiments using three-color analysis with FITC, PE, and avidin to determine the proportion 40 - , , of CD3+,CD4-,CD8~ and Leu 19+, CD3~,CD16" cells were in 30 - accordance with results obtained in six experiments where these 20 â€¢¿ populations were determined by 2-color staining of the various 10 - subpopulations.

10:1 3:1 1:1 70 r K562 DISCUSSION 60 â€¢¿ We have recently reported the development of long-term (14 50 to 21 days) cultures of cells mediating LAK activity using 40 stimulation with OKT3 + IL2. The large increase in cell number 30 induced by this form of culture and the presence of high level 20 LAK activity at 14 days suggested that (a) some of the prolif 10 â€¢¿ V erating cells were the same as those mediating LAK activity in 0 10:1 3:1 1:1 .3:1 short-term cultures and/or that (Â¿>)othercells, not shown to Effector: Target manifest LAK function in short-term cultures, were activated Fig. 6. Day 13 PBL were depleted of CD16, CD4, and CDS cells. Remaining to mediate LAK activity. In fact, it appears that both of these cells were sorted with PE-Leu 19 and FITC-Leu 4 (CD3). LAK activity was tested immediately after sorting. Undepleted PBL were used as control. factors are operative. The initial phase of isolation of cells mediating LAK activity Table 2 Average CD}' T-cell and CO/6* NK cell increases in OKT3 + demonstrated that both CD3+,CD16~ cells and CD16+,CD3~ Â¡L2-stimulated cultures cells developed essentially equivalent LAK activity in these Cell no. x 10" long-term cultures. This finding contrasted with the results DayO Day 14 Fold expansion showing that LAK function in short-term cultures was mainly PBL +IL2*CD3*CD16*PBL expressed by CD3~ cells, i.e., CD16 and/or Leu 19+. LAK activity of the CD3+ cells could potentially be attributed to the presence of residual OKT3 on the surface of the cells after +OKT3*CD3*CD positive selection. Experiments where CD3+ cells were enriched by negative selection, i.e., depletion of CD16+,CD3~ cells, 16*2.30.42.30.43072063513189088 yielded highly enriched CD3+ populations which demonstrated " Cell numbers were calculated based on the percentage of positive cells for Leu 4 (CD3) or Leu 11 (CD 16) moAbs as determined by fluorescence analysis. equivalent activity to cells positively selected by sorting with Total number of cells was determined by cell counts every 48 h. OKT3 (data not shown). These findings suggested that the LAK 'Total expression of PBL + IL2 in 14-day culture was 9.8-fold, while cells activity present in the CD3+ cells was not due to binding stimulated with OK I 3 increased 49-fold. between the target and effector cells induced by the residual Table 3 Percentage of CAT,CD4~,CD8~ and LeuW,CD3~,CD16~ cells present OKT3. Binding and lysis of tumor targets, though, has been in long-term culture with Â¡LÃ¬+OKT3 reported using heteroconjugates between anti-CD3 and tumor- CD3*,CD8*Â°CD3*,CD4*Â°CD3*,CD4-,CD8~* Leul9*,CD3-,CD16~* specific moAbs (16). Further dissection identified two other cell subpopulations, CD3+,CD4",CD8" cells and Leu DayO 39 51 7 3 Day 14 53 10 24 13 19+,CD3~,CD16~ cells, which also developed LAK activity in Â°Double-color analysis was done using FITC- or PE-conjugated moAb Leu 4 long-term cultures with OKT3 + IL2. (CD3), Leu 2 (CDS), and Leu 3 (CD4). LAK activity of CD3+,CD16- cells and CD16+,CD3- cells '' Three-color analysis was done using biotin-conjugated Leu 4 (CD3) and FITC- or PE-conjugated Leu 2 (CDS), Leu 3 (CD4), Liu 11 (CD 16), and Leu 19. was higher than unseparated PBL in 4 of 9 experiments and equivalent in the other 5 experiments. It seems likely that, in promote the growth and proliferation of CD3+ T-cells, but also, those cases where isolated subpopulations showed much higher and to essentially the same magnitude, the proliferation of LAK activity than unseparated PBL, this reflects enrichment CD16+ NK cells. In six experiments when PBL were cultured of cells mediating LAK activity or that isolation of the different in IL2 for 14 days, CD3+ cells showed an average 32-fold subpopulations eliminates regulatory mechanisms otherwise increase in number, while CD16+ cells increased an average of active in unseparated cells. In contrast, other purified subpopulations, i.e., CD3+,CD4-,CD8- and Leu 19+,CD3-,CD16-, 34 times their original number. Results from eight experiments where cells were stimulated with OKT3 + IL2 showed an showed LAK activity similar to and not higher than unseparated increase in cell number of both CD3+ T-cells (154-fold) and PBL. We do not have data which can explain this finding; one CD 16* NK cells (average, 128-fold). Cells coexpressing CD3 possible explanation would be that CD16+,CD3" cells have higher LAK on a per cell basis than CD3+,CD4",CD8" or Leu and CD16 were <5% throughout the culture period in all 19+,CD3~,CD16~ cells. Alternatively, LAK function in the bulk experiments. 966

population might be the result of synergistic interaction be vation of NK cells have not been as well determined as those tween several subpopulations. for T-cells, it has been demonstrated that, in order to maintain Development of NK activity by T-cells under certain condi NK function in clones, these cells require the continuous pres tions has been previously reported. Reports have shown that ence of IL2 and feeder cell monolayers with cells such as K562 cloned murine cytotoxic T-cells were able to develop non- or lymphoblastoid cell lines (29, 30). It seems unlikely that MHC-restricted lytic activity (promiscuous killing) when cul OKT3 would directly activate cells which cannot bind this tured in supernatants derived from mitogen-stimulated PBL moAb as determined by immunofluorescence. It is most prob (17-19). This observation could be explained in part by the able, therefore, that the secretion of multiple lymphokines from presence of factors such as IFN-a and IFN-/3, which were the different subpopulations being activated by OKT3 + IL2 subsequently shown to be able to induce Nu-type function in results in a suitable environment for enhanced proliferation of antigen-specific CTL (20, 21). It seemed likely that, as a result NK cells. That the proliferation of CD16+ NK cells may be due of the polyclonal nature of activation with OKT3 + IL2, factors to production of factors by activated T-cells is suggested by the such as IFN would be secreted into the culture medium with initial decrease in the proportion of NK cells at a time when the resulting development of non-MHC-restricted lysis by CD3* T-cells already show a significant increase in number, CD3+ CTL. However, in our experiments, it was primarily the but may not yet be activated maximally in terms of providing CD3+,CD4-,CD8~ cells within the CD3+ cells that developed lymphokines for the growth of CD16+ cells. This in turn is high LAK activity, while CD3+,CD8+ or CD3+,CD4+ cells followed, several days later, by an increase in the number of developed much lower LAK function. The CD3+,CD4~,CD8~ CD 16+NK cells. CD3+,CD4~,CD8~ cells could be activated by the direct bind cells isolated did not bind moAb WT31, which identifies the aÃŸheterodimer information of the T-cell receptor. These cells ing of OKT3, and as such showed a significant increase from have instead been shown to express a yo heterodimer or a yy 7% to 24% of the total population in the experiment presented homodimer configuration of the T-cell receptor (22-24). Func and, in some experiments, as high as 30% (data not shown). In tionally, these cells express NK activity which can be enhanced contrast to CD3+,CD8+ cytotoxic T-cells, where the different by the addition of IFN-/3 (22). It is therefore not final as to signals leading to the development of cytotoxicity have been whether these are T-cells because of the expression of CD3 or established, only preliminary information is known as to the signals necessary to stimulate CD3+,CD4~,CD8~ cells to pro NK cells due to their functional characteristics. It has recently been shown that antibody directed against this novel yo T-cell liferate and develop LAK activity (31). Whether IL2 and OKT3 receptor blocks lytic function, suggesting that this structure serve as separate differentiation or proliferation signals for serves as a functional receptor (24). CD3*,CD4~,CD8~ cells these cells is yet to be established. isolated in the present experiments did not coexpress Leu 19 which contrasts with other reports (24). Interestingly, Bailas ACKNOWLEDGMENTS and Rasmussen (25) recently demonstrated that murine Lyt2~,L3t4~ thymocytes had significantly higher LAK activity than did Lyt2*,L3t4" (cytotoxic/suppressor) and Lyt2~,L3t4+ The authors would like to thank Nancy Andresen, Diane Buswell, and Connie Greenberg for preparing the manuscript, Beth Kovacs for (helper) subpopulations. It still remains to be determined technical assistance and Brad Anderson for running the fluorescence- whether the in vitro antitumor activity of the CD3+,CD4",CD8" activated cell sorter. cells has an antitumor role in vivo. Preliminary information by Mule et al. (26) has shown a role for similar cells in in vivo REFERENCES antitumor responses in mice treated only with infusions of IL2. As reported previously by others, CD3+,CD8+ and CD3+,CD4+ 1. Zarling, J. M., Kay, N. E., Grant, B., Yasukawa, M., and Bach F. H. Human cells developed much lower LAK activity than the other sub- anti-lymphoma responses generated in vitro and in vivo following sensitization with allogeneic leukocytes. Cancer Immunol. Immunother., IS: 237- populations, although in one of five experiments the LAK 239, 1983. function was significant (Fig. 2). 2. Zarling. J. M.. Robins. H. I.. Raich, P. C, Bach, F. H., and Bach, M. L. Generation of cytotoxic T-lymphocytes to autologous human leukaemia cells Recent reports by Lanier et al. (27) have shown the presence by sensitization to pooled allogeneic normal cells. Nature (Lond.), 274:269- of a subpopulation expressing NK activity phenotypically dis 271, 1978. tinct from CD16+,Leu 19+ cells. These cells are Leu 19+,CD2+ 3. Ini/r, M., Grimm, E. A., Mazumder, A., Strausser, J. L. and Rosenberg, S. A. Lysis of fresh and cultured autologous tumor by human lymphocytes but do not express either CD3 or CD16. It seems unlikely that cultured in T-cell growth factor. Cancer Res., 41:4420-4425, 1981. the lack of expression of CD3 can be explained by modulation 4. Grimm, C., Mazumder, A., Zhang, H., and Rosenberg, S. Lymphokine- produced by the use of OKT3 for the initial stimulation or for activated killer phenomenon: lysis of natural killer-resistant fresh solid tumor cells by interleukin-2 activated autologous human peripheral blood lympho the cell separation. The OKT3 used for stimulation was put in cytes. J. Exp. Med., 155: 1823-1841, 1982. the culture only on the first day [while cells were separated 12 5. Grimm, E. A., Gorelik, E., Rosenstein, M. M., and Rosenberg, S. The days later (Day 13)], and the separation procedure is done at lymphokine-activated killer cell phenomenon: in vitro and in vivo studies. In: 4Â°C,greatly minimizing the possible modulation of CD3 during S. Cohen and J. Oppenheim (eds.), Interleukins, Lymphokines, and Cyto- kines, pp. 739-748, New York: Academic Press, 1983. this procedure. It has also been reported that cells with similar 6. Ochoa, A. C., Gromo, G., Alter, B. J., Sonde!, P. M., and Bach, F. H. Long- term growth of lymphokine activated killer (LAK) cells: role of anti-CD3, 0- phenotypic characteristics do not express a functional T-cell 1L1, Â¡nterferony and lÃ¬.J.Immunol., 138: 2728-2733, 1987. receptor, but instead express a 1.0-kilobase truncated Ti ÃŸ 7. Grimm, E. A., Ramsey, K. M., Mazumder, A., Wilson, D. J., Djeu, J., and transcript without a Ti a transcript of the T-cell receptor (28). Rosenberg, S. Lymphokine activated killer cell phenomenon: II precursor phenotype is serologically distinct from peripheral T lymphocytes, memory Data reported here demonstrate that this subpopulation also cytotoxic thymus-derived lymphocytes, and natural killer cells. J. Exp. Med., develops high LAK activity and increases in number in the 15 7: 884-897, 1983. 8. Ortaldo. J. R., Mason, A., and Overton, R. Lymphokine-activated killer cells. long-term cultures stimulated with OKT3 + IL2. Analysis of progenitors and effectors. J. Exp. Med., 164: \ 193-1205, 1986. The increase in cell number induced by OKT3 -I-IL2 stimu 9. Ferrini, S., Miescher, S., Zocchi, M. R., Fliender, V., and Moretta, A. lation was not limited to the CD3+ population. Interestingly, Phenotypic and functional characterization of recombinant interleukin-2 CD16+ NK cells showed a parallel increase in cell number to (HI J) indiimi activated killer cells: analysis at the population and clonal levels. J. Immunol., 138: 1297-1302, 1987. CD3+ T-cells. Although the signals for proliferation and acti 10. Itoh, K., Tilden, A. B., and Balch, C. M. Lysis of human solid tumor cells 967

by lymphokine-activated natural killer cells. J. Inumino!.. 136: 3910-3915, Ronteltap, C. P. M.. and Bolhuis. R. L. H. Lysis of tumor cells by CD3*4~8~16* T cell receptor of aÃŸ~clones, regulated via CD3 and CD 16 1986. 11. Phillips, J. H., and Lanier, L. L. Dissection of the lymphokine-activated activation sites, recombinant interleukin 2, and interferon .1. J. Immunol., killer phenomenon. Relative contribution of peripheral blood natural killer 138: 1627-1633, 1987. cells and T lymphocytes to cytolysis. J. Exp. Med., 164: 814-825, 1986. 23. Van de Griend, R., van Krimpen, B., Ronteltap, C. P. M., and Bolhuis, R. 12. Wang, A., Lu, S. D., and Mark, D. Site specific mutagenesis of the human L. H. Rapidly expanded activated human killer cell clones have strong anti- interleukin-2 gene: structure-function analysis of the cysteine residues. Sci tumor activity and have the surface phenotype of either T>, T-non-y, or null ence (Wash. DC), 224: 1431-1433, 1984. cells. J. Immunol., 132: 3185-3194, 1984. 13. Rosenberg, S. A., Grimm. E. A., McGrogan, M.. Doyle, M., Kawasaki. E., 24. Moingeon, P., Jitsukawa, J., Faure, F., Troalen, F., Triebel, F., Graziani, Koths. K.. and Mark, D. F. Biological activity of recombinant interleukin-2 M., Forestier, F., Bellet, D., Bohvon, C., and Hercend, T. A 7 chain complex produced in Escherichia coli. Science (Wash. DC), 223: 1412-1415, 1984. forms a functional receptor on cloned human lymphocytes with natural killer-like activity. Nature (Lond.), 325: 723-726, 1987. 14. Wee, S. L., Wu, S., Alter, B. J., and Bach, F. H. Early detection and specificity 25. Bailas, Z. K., and Rasmussen, W. Lymphokine-activated killer (LAK) cells. analysis of human cytolytic T lymphocytes (CTL) colonies generated in soft agarose culture: a potential assay for definition of CTL defined (CD) deter III. Characterization of LAK precursors and susceptible target cells within minants. Hum. Inumino!.. 3:45-56, 1981. the murine thymus. J. Immunol., 139: 3542-3549, 1987. 15. Wysocki, L. J., and Sato, V. L. "Panning" for lymphocytes: a method for 26. Mule, J. J., Yang, J. C., Lafreniere, R., Shu, S., and Rosenberg, S. A. cell selection. Proc. Nati. Acad. Sci. USA, 75: 2844-2848. 1978. Identification of cellular mechanisms operational in vivoduring the regression of established pulmonary mÃ©tastasesbythe systemic administration of high- 16. Perez, P., Hoffman. R. W., Shaw. S., Bluestone, J. A., and Segal, D. M. dose recombinant interleukin-2. J. Immunol., 139: 285-294, 1987. Specific targeting of cytotoxic T cells by anti-T3 linked to anti-target cell 27. Lanier, L. L., Le, A. M., Civin, C. I., Loken, M. R., and Phillips, J. H. The antibody. Nature (Lond.), 316: 354-356, 1985. relationship of CD 16 (Leu 11) and Leu 19 (NKII-1) antigen expression on 17. Shortman, K., Wilson, A., Scollay, R.. and Chen, W. F. Development of human peripheral blood NK cells and cytotoxic T lymphocytes. J. Immunol., large granular lymphocytes with anomalous, nonspecific cytotoxicity in clone derived from Ly-2* T cells. Proc. Nati. Acad. Sci. USA, 80: 2728-2732, 136:4480-4486, 1986. 28. Ritz, J., Schmidt, R. E., Royer, H. D., Hercend, T., Hussey, R. E., and 1983. Reinherz, E. L. Analysis of T cell receptor gene rearrangements and expres 18. Havele, C., Bleackley. R. C., and Paetkav, V. Conversion of specific to sion in human natural killer clones. Science (Wash. DC), 22S: 1540-1543, nonspecific cytotoxic T lymphocytes. J. Immunol., 137: 1448-1454, 1986. 1985. 19. Kane, K. P.. and Clark. W. R. Class I MHC antigens in the generation and 29. Olabuenaga, S. E., Brooks, C. G., Gillis, S., and Henney, C. S. Interleukin 2 expression of promiscuous cytotoxic cell function. J. Immunol., 137: 3080- is not sufficient for the continuous growth of cloned NK-like cytotoxic cell 3086. 1986. lines. J. Immunol., 131: 2386-2391, 1983. 20. Brooks, C. G. Reversible induction of natural killer cell activity in cloned 30. Hercend, T., Reinherz, E. L., Meuer, S. C., Schlossman, S. F., and Ritz, J. murine cytotoxic T lymphocytes. Nature (Lond.). 305: 155-158, 1983. Phenotypic and functional heterogeneity of human cloned natural killer cell 21. Brooks, C., Holschen, M., and Urdal, D. Natural killer activity in cloned lines. Nature (Lond.), 301: 158-160, 1983. cytotoxic lymphocytes: regulation by interleukin 2, interferon, and specific 31. DeLaÃ±ara,A., Toribio, M. K., Marquez, C., and Martinez, C. A. Interleukin antigen. J. Immunol.. 135: 1145-1152. 1985. 2 promotes growth and cytolytic activity in human T3*4~8~ thymocytes. 22. van de Griend, R. J., Tac, W. J. M., van Krimpen, B. A., Breugdenhil, R. J., Proc. Nati. Acad. Sci. USA, 82:6268-6271, 1985.

968

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1989 American Association for Cancer Research. Lymphokine-activated Killer Activity in Long-Term Cultures with Anti-CD3 plus Interleukin 2: Identification and Isolation of Effector Subsets

Augusto C. Ochoa, Diane E. Hasz, Rebecca Rezonzew, et al.

Cancer Res 1989;49:963-968.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/49/4/963

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/49/4/963. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.