'0022-5347 /92/1471-0207:i"W3.00/0 THE ,JOURNAL OF UROLOGY Copyright@ 1992 by AMERICAN UROLOGICAL ASSOCIATION, INC.
Vol. 147, 207-211, January 1992 Printed in U.S.A.
AN INVESTIGATION OF FACTORS INFLUENCING THE IN VITRO INDUCTION OF LAK ACTIVITY AGAINST A VARIETY OF HUMAN BLADDER CANCER CELL LINES ANDREW M. JACKSON,* SIMON J. HAWKYARD, STEVEN PRESCOTT, ALASTAIR W. S. RITCHIE, KEITH JAMES AND GEOFFREY D. CHISHOLM From the Departments of Surgery/Urology, Western General Hospital, Edinburgh, Scotland
ABSTRACT We investigated the sensitivity of transitional cell carcinoma cells, derived from the human bladder, to lymphokine activated killer cells. Recombinant interleukin-2 activated peripheral blood mononuclear cells were studied for their ability to mediate the cytolysis of a panel of four established human bladder transitional cell carcinoma cell lines. Lymphokine activated killer activity was assessed using a standard four hour chromium release assay. All four bladder cancer cell lines proved to be susceptible to lymphokine activated killer mediated cytolysis. This was found to be dependent upon the dose of cytokine and upon the duration of the activation period. The four cell lines were differentially susceptible to lysis (specific cytotoxicity at effector to target ratio of 40:1; RT112 = 22.9%, RT4 = 49.2%, MGH-Ul = 49.1%, EJ18 = 62.3%). The varying susceptibility of lymphokine activated killer mediated cytotoxicity was found to be independent of the histological grade of the parent tumour or the donor of effector cells. Both interferon-alpha and tumour necrosis factor-alpha also elicited lymphokine activated killer cell activity, although the maximum specific cytotoxicity achieved was considerably lower than that obtained with interleukin-2 alone. Interleukin-2, at optimal concentration, and tumour necrosis factor-alpha were found to behave synergistically in the generation of lymphokine activated killer effectors. However, concentrations of tumour necrosis factor-alpha higher than 100 Uml.- 1 resulted in a decrease in specific cytotoxicity. These findings suggest a possible use of adoptive immunotherapy in human bladder cancer and indicate the optimum conditions for the generation of such effector cells. KEY WORDS: bladder neoplasms; immunotherapy, adoptive
The stimulation of peripheral blood lymphocytes with recom binant interleukin-2 (IL-2) results in the generation of lym phokine activated killer (LAK) cells. These cells are distin guishable from conventional natural killer (NK) cells by their ability to lyse otherwise NK resistant tumour cell targets. 1-4 This cytolytic activity is distinct from T cell mediated killing in that it is non-MHC restricted1 and does not affect normal cells. 5 IL-2 has been attributed with the ability of generating LAK cells capable of tumour cytolysis. However, other cytokines have also been found to be involved in the generation of LAK activity. Using neutralising antibodies against interferon gamma (IFN-y ) and IL-2 a partial inhibition of LAK activity was noted indicating that IFN-y plays a functional role in the optimum induction of LAK cells by IL-2. 6 Tumour necrosis factor-alpha (TNFa), interleukin-I (IL-1), and interleukin-4 (IL-4) have also been shown to partake in the development of LAK cells. 7-9 TNFa was found to act synergistically with sub optimal concentrations of IL-2 in the induction of LAK activ ity, however, no increase above that achieved with optimal concentrations of IL-2 alone was reported. Similar results were to be found when the actions of IL-1 on LAK precursors were investigated. IL-4 has been shown to negatively regulate IL-2induced LAK cells; 1° however, TNFa was found to partially overcome the inhibitory effects of IL-4. 7 The mechanisms by which LAK cells mediate the cytolysis of their targets remain largely unknown. It is, however, likely that some of the soluble Accepted for publication June 25, 1991. * Requests for reprints: Department of Surgery, University of Edin burgh Medical School, Teviot Place, Edinburgh EH8 9AG, Scotland, United Kingdom. Supported by the Cancer Research Campaign of Great Britain.
factors shown to be released by LAK cells (such as IFN-y , TNFa, and IL-la11· 12) are involved in target cell lysis. The precursors of the LAK cell have been the subject of some controversy. The debate surrounds the issue of whether these precursors are NK-like, T cell-like or neither. 2· 13• 14 It would seem that a heterogeneous population of cells is responsible for LAK activity and that this population of cells varies between individual donors. However, the majority of LAK activity gen erated by IL-2 would seem to be mediated by cells expressing the CD56 antigen and with the morphology of large granular lymphocytes. In this study, we have attempted to determine the phenotype of LAK populations following in vitro activation with recombinant IL-2. The majority of LAK cell studies to date have investigated non-urological malignancies such as malignant melanoma, co lorectal cancer and breast cancer. Those studies with a urolog ical interest have tended to concentrate on renal cell carcinoma. A recent study by Rosenberg et al. 3 examin�d the ability of tumour infiltrating lymphocytes to lyse fresh autologous nonrenal urological malignancies. The results showed that the majority of tumour infiltrating lymphocytes (TIL) were lytic against both autologous and allogeneic targets. The accessibil ity of the bladder tends itself to intravesicle instillation and intralesional administration of LAK cells. To investigate the potential in vivo application of peripheral blood LAK cells the sensitivity of long term transitional cell carcinoma (TCC) cell lines was examined using a standard cytotoxicity assay. We demonstrate that four bladder tumour cell lines are differentially susceptible to LAK mediated cytol ysis. These studies were performed to ascertain optimal condi tions for generating LAK cells and to provide information on the mechanisms involved.
207
208
JACKSON AND ASSOCIATES TABLE 1. Sources and specificities of monoclonal antibodies used for
MATERIALS AND METHODS
phenotypic studies on LAK cells
Effector cells. Mononuclear cells were isolated from fresh,
heparinized peripheral blood from normal, healthy laboratory volunteers. Briefly, peripheral blood was centrifuged on Ficoll Isopaque (density= 1.086 gm./cm.3 ) for 30 min at 400 g. The interface was collected and subsequently washed three times in Hanks Balanced Salt Solution (HESS) after which adherent cells were removed by adherence to plastic for one hour at 37C. The non-adherent cells were suspended at a density of 2 x 106 ml.- 1 in foetal calf serum (FCS) containing 10% dimethylsul phoxide (DMSO) and stored in liquid nitrogen until required. Bladder cancer cell lines. Human TCC cell lines RT4, RT112 & MGH-Ul (corresponding to histological grades Gl, G2 and G3 respectively) were kindly provided by Dr. J. Masters of the Instit-ute 0f Urology, London. EJ18, a sub-line of similar gen otype to MGH-Ul, was the gift of Dr. J. Boyd. All were tested mycoplasma free by the Public Health Laboratory Service, Porton Down, England. Cells were seeded at 5 X 104 m1.- and cultured in antibiotic free RPMI 1640 (GIBCO) containing 5% FCS in 25 cm.2 tissue culture flasks (J. Bibby Science Products Ltd., Stone, Staffs., England). Cells were recovered for routine use in cytotoxicity assays by trypsinisation (trypsin/EDTA, 0.5 gl.- trypsin and 0.2 gl.- disodium EDTA). Source of cytokines. Highly purified recombinant IL-2 was generously provided by Euro-Cetus B.V. (The Netherlands) and contained 1.8 X 107 Umg.- of protein. Recombinant IFN, and TNF a were purchased from Boehringer Mannheim and contained 2.5 x 107 Umg.- and 2.0 X 107 Umg.- of protein respectively. Recombinant IFNa was obtained from Dr. S. Langden, (ICRF, Dept. Oncology, University of Edinburgh). Generation of LAK cells. Mononuclear cells were cultured at 1 X 106 ml.- 1 in RPMI 1640 medium supplemented with 5% heat inactivated FCS and varying concentrations of recombi nant interleukin-2. The mononuclear cells were incubated at 37C in a humidified atmosphere of 5% CO2 in air for up to six days. After the incubation the cells were washed twice in HESS and resuspended at 4 X 106 ml.- 1 and assessed for cytolytic activity against labelled target cells. The presence of FCS in the medium was not found to influence the generation of LAK cells as compared to serum free medium (data not shown). Cytotoxicity assay. Cytotoxicity was measured by means of a four-hour chromium-release assay. In brief 106 target cells were radiolabelled with 100 µCi Na2 Cr0 4 (Amersham Interna tional, Amersham, UK.) at 37C for one hour with shaking. After multiple washes the targets were resuspended in RPMI 1640 at a density of 5 X 104 ml.- 1 Labelled targets (5 x 103 well- 1) were dispensed into the wells of 96-well U-bottomed microtitre plates (Sterilin Ltd., Feltham, England) and then graded numbers of effector cells were added to give effector-to target ratios (E:T) ranging from 40:1 to 1.25:1 in a total of 200 µl. of medium. After a four-hour incubation at 37C and cen trifugation at 100 g for five minutes, 100 µl. of supernatant was removed from each well and the radioactive content determined using a liquid scintillation counter (Canberra Packard Ltd.). All assays were performed in triplicate and the data were calculated from E:T of 40:1 in order to allow comparison between lines of differing susceptibility and expressed as the percentage specific cytotoxicity, calculated as below;
Monoclonal Antibody
Specificity
Source
CD3
T lymphocytes
BD
CD4
Helper T cells
SAPU
CD8
Cytotoxic T cells
SAPU
CD16
NK cells
BD
IL-2R
Receptor for interleukin-2
Amersham
HNK-1
Fe receptor on NK cells
BD
BD - Beclon Dickenson Ltd. (Cowley, UK) SAPU - Scottish Antibody Production Unit (Carluke, UK) Am-Amersham International (UK)
1
1
1
1
1
1
51
%Specific Cytotoxicity cpm exp release - cpm spontaneous release cpm total release - cpm spontaneous release Source of monoctonal antibodies. For details of antibody spec ificity and source see table 1. Monoclonal antibodies which were not directly conjugated to a fluorochrome were detected using a goat anti-mouse-fluorescein isothiocyanate (FITC) con jugate (Sigma Chemical Co., USA). Fluorescence staining and flow cytometric analysis. Two col our immunofluorescence was performed on effector cells
stained with monoclonal antibodies conjugated to either FITC or phycoerythrin. For each analysis 5 x 105 cells were washed with washing buffer (PBS, 1% FCS, 0.01% sodium azide) and incubated at 4C for 30 minutes with optimal concentrations of primary monoclonal antibody. Following two washes, the bind ing of unconjugated monoclonal antibodies was detected using a goat anti-mouse-FITC conjugate for a further 30 minutes at 4C. The cells were then blocked with a 10% (v/v) solution of normal mouse serum for 30 minutes and, after washing, incu bated with a second monoclonal antibody conjugated to phy coerythrin for a further 30 minutes. The cells were washed twice more and resuspended in 1% formaldehyde prior to analysis using a flow micro-fluorometer (Coulter, EPICS-C, USA). Non-viable cells were gated out of the window and at least 10,000 events were accumulated by using logarithmic amplification of fluorescence intensity. Statistical analysis. All experiments were performed a mini mum of five times in triplicate. Statistical analysis was per formed using the non-para-metric "Mann-Whitney U" test. The data were analysed using Statview 512 software on an Apple Macintosh Hex computer. RESULTS
Dose dependent kinetics of LAK cell induction. Although previous studies have demonstrated that IL-2 is capable of generating from PBMC LAK cells directed against a variety of tumour cells, little work has been undertaken on non-renal urological malignancies and in particular TCC of the bladder. In order to evaluate the ability of LAK cells to mediate cytolytic activity against bladder cancer, normal PBL were cultured with increasing doses of IL-2 for up to 10 days. The results clearly demonstrate that IL-2 can induce PBL to exhibit LAK activity against a panel of bladder cancer cell lines (figure 1). The induction of LAK activity was found to be dependent upon the dose of IL-2 and the duration of the activation period. As little as 10 Uml.- rIL-2 was able to induce LAK activity against all four cell lines after only three days activation. However, optimal levels of LAK activity were achieved after six days incubation with 1,000 Uml.- IL-2. Any further increase in the dose of IL-2 or in the duration of the activation period failed to result in a corresponding increase in lytic potential (data not shown). The four cell lines investigated were not killed with equal efficiency by the same population of effector cells. The grade 2 tumour, RT112, was least susceptible to LAK mediated cytol ysis (mean± standard error, 22.9%± 2.5), RT4 (Gl) (49.2%± 6.0) and MGH-Ul (G3) (49.1% ± 3.9) were more susceptible, and finally EJ18 (G3) was most susceptible (62.3%± 3.9). Influence of PBMC donor on LAK killing. The cytotoxic potential of LAK cells derived from five donors was simulta neously assessed against the RT112 and MGH-Ul cell lines. The results, shown in figure 2, demonstrate slight differences 1
1
209
LAK ACTIVITY AGAINST BLADDER TUMOUR CELLS IN VITRO
..
�
:�
��
u (J
,.:: ·i:;
.,...
.."....
00
i:,.
·�1
Ill
� l!lil �
RT4
,.
100 ulL-2/Mi 1000 aBL-2/ml
�_J
20
•
.."'
100
� 40 hours
i....
[:2l
"
Vol. 147, 207-211, January 1992 Printed in U.S.A.
AN INVESTIGATION OF FACTORS INFLUENCING THE IN VITRO INDUCTION OF LAK ACTIVITY AGAINST A VARIETY OF HUMAN BLADDER CANCER CELL LINES ANDREW M. JACKSON,* SIMON J. HAWKYARD, STEVEN PRESCOTT, ALASTAIR W. S. RITCHIE, KEITH JAMES AND GEOFFREY D. CHISHOLM From the Departments of Surgery/Urology, Western General Hospital, Edinburgh, Scotland
ABSTRACT We investigated the sensitivity of transitional cell carcinoma cells, derived from the human bladder, to lymphokine activated killer cells. Recombinant interleukin-2 activated peripheral blood mononuclear cells were studied for their ability to mediate the cytolysis of a panel of four established human bladder transitional cell carcinoma cell lines. Lymphokine activated killer activity was assessed using a standard four hour chromium release assay. All four bladder cancer cell lines proved to be susceptible to lymphokine activated killer mediated cytolysis. This was found to be dependent upon the dose of cytokine and upon the duration of the activation period. The four cell lines were differentially susceptible to lysis (specific cytotoxicity at effector to target ratio of 40:1; RT112 = 22.9%, RT4 = 49.2%, MGH-Ul = 49.1%, EJ18 = 62.3%). The varying susceptibility of lymphokine activated killer mediated cytotoxicity was found to be independent of the histological grade of the parent tumour or the donor of effector cells. Both interferon-alpha and tumour necrosis factor-alpha also elicited lymphokine activated killer cell activity, although the maximum specific cytotoxicity achieved was considerably lower than that obtained with interleukin-2 alone. Interleukin-2, at optimal concentration, and tumour necrosis factor-alpha were found to behave synergistically in the generation of lymphokine activated killer effectors. However, concentrations of tumour necrosis factor-alpha higher than 100 Uml.- 1 resulted in a decrease in specific cytotoxicity. These findings suggest a possible use of adoptive immunotherapy in human bladder cancer and indicate the optimum conditions for the generation of such effector cells. KEY WORDS: bladder neoplasms; immunotherapy, adoptive
The stimulation of peripheral blood lymphocytes with recom binant interleukin-2 (IL-2) results in the generation of lym phokine activated killer (LAK) cells. These cells are distin guishable from conventional natural killer (NK) cells by their ability to lyse otherwise NK resistant tumour cell targets. 1-4 This cytolytic activity is distinct from T cell mediated killing in that it is non-MHC restricted1 and does not affect normal cells. 5 IL-2 has been attributed with the ability of generating LAK cells capable of tumour cytolysis. However, other cytokines have also been found to be involved in the generation of LAK activity. Using neutralising antibodies against interferon gamma (IFN-y ) and IL-2 a partial inhibition of LAK activity was noted indicating that IFN-y plays a functional role in the optimum induction of LAK cells by IL-2. 6 Tumour necrosis factor-alpha (TNFa), interleukin-I (IL-1), and interleukin-4 (IL-4) have also been shown to partake in the development of LAK cells. 7-9 TNFa was found to act synergistically with sub optimal concentrations of IL-2 in the induction of LAK activ ity, however, no increase above that achieved with optimal concentrations of IL-2 alone was reported. Similar results were to be found when the actions of IL-1 on LAK precursors were investigated. IL-4 has been shown to negatively regulate IL-2induced LAK cells; 1° however, TNFa was found to partially overcome the inhibitory effects of IL-4. 7 The mechanisms by which LAK cells mediate the cytolysis of their targets remain largely unknown. It is, however, likely that some of the soluble Accepted for publication June 25, 1991. * Requests for reprints: Department of Surgery, University of Edin burgh Medical School, Teviot Place, Edinburgh EH8 9AG, Scotland, United Kingdom. Supported by the Cancer Research Campaign of Great Britain.
factors shown to be released by LAK cells (such as IFN-y , TNFa, and IL-la11· 12) are involved in target cell lysis. The precursors of the LAK cell have been the subject of some controversy. The debate surrounds the issue of whether these precursors are NK-like, T cell-like or neither. 2· 13• 14 It would seem that a heterogeneous population of cells is responsible for LAK activity and that this population of cells varies between individual donors. However, the majority of LAK activity gen erated by IL-2 would seem to be mediated by cells expressing the CD56 antigen and with the morphology of large granular lymphocytes. In this study, we have attempted to determine the phenotype of LAK populations following in vitro activation with recombinant IL-2. The majority of LAK cell studies to date have investigated non-urological malignancies such as malignant melanoma, co lorectal cancer and breast cancer. Those studies with a urolog ical interest have tended to concentrate on renal cell carcinoma. A recent study by Rosenberg et al. 3 examin�d the ability of tumour infiltrating lymphocytes to lyse fresh autologous nonrenal urological malignancies. The results showed that the majority of tumour infiltrating lymphocytes (TIL) were lytic against both autologous and allogeneic targets. The accessibil ity of the bladder tends itself to intravesicle instillation and intralesional administration of LAK cells. To investigate the potential in vivo application of peripheral blood LAK cells the sensitivity of long term transitional cell carcinoma (TCC) cell lines was examined using a standard cytotoxicity assay. We demonstrate that four bladder tumour cell lines are differentially susceptible to LAK mediated cytol ysis. These studies were performed to ascertain optimal condi tions for generating LAK cells and to provide information on the mechanisms involved.
207
208
JACKSON AND ASSOCIATES TABLE 1. Sources and specificities of monoclonal antibodies used for
MATERIALS AND METHODS
phenotypic studies on LAK cells
Effector cells. Mononuclear cells were isolated from fresh,
heparinized peripheral blood from normal, healthy laboratory volunteers. Briefly, peripheral blood was centrifuged on Ficoll Isopaque (density= 1.086 gm./cm.3 ) for 30 min at 400 g. The interface was collected and subsequently washed three times in Hanks Balanced Salt Solution (HESS) after which adherent cells were removed by adherence to plastic for one hour at 37C. The non-adherent cells were suspended at a density of 2 x 106 ml.- 1 in foetal calf serum (FCS) containing 10% dimethylsul phoxide (DMSO) and stored in liquid nitrogen until required. Bladder cancer cell lines. Human TCC cell lines RT4, RT112 & MGH-Ul (corresponding to histological grades Gl, G2 and G3 respectively) were kindly provided by Dr. J. Masters of the Instit-ute 0f Urology, London. EJ18, a sub-line of similar gen otype to MGH-Ul, was the gift of Dr. J. Boyd. All were tested mycoplasma free by the Public Health Laboratory Service, Porton Down, England. Cells were seeded at 5 X 104 m1.- and cultured in antibiotic free RPMI 1640 (GIBCO) containing 5% FCS in 25 cm.2 tissue culture flasks (J. Bibby Science Products Ltd., Stone, Staffs., England). Cells were recovered for routine use in cytotoxicity assays by trypsinisation (trypsin/EDTA, 0.5 gl.- trypsin and 0.2 gl.- disodium EDTA). Source of cytokines. Highly purified recombinant IL-2 was generously provided by Euro-Cetus B.V. (The Netherlands) and contained 1.8 X 107 Umg.- of protein. Recombinant IFN, and TNF a were purchased from Boehringer Mannheim and contained 2.5 x 107 Umg.- and 2.0 X 107 Umg.- of protein respectively. Recombinant IFNa was obtained from Dr. S. Langden, (ICRF, Dept. Oncology, University of Edinburgh). Generation of LAK cells. Mononuclear cells were cultured at 1 X 106 ml.- 1 in RPMI 1640 medium supplemented with 5% heat inactivated FCS and varying concentrations of recombi nant interleukin-2. The mononuclear cells were incubated at 37C in a humidified atmosphere of 5% CO2 in air for up to six days. After the incubation the cells were washed twice in HESS and resuspended at 4 X 106 ml.- 1 and assessed for cytolytic activity against labelled target cells. The presence of FCS in the medium was not found to influence the generation of LAK cells as compared to serum free medium (data not shown). Cytotoxicity assay. Cytotoxicity was measured by means of a four-hour chromium-release assay. In brief 106 target cells were radiolabelled with 100 µCi Na2 Cr0 4 (Amersham Interna tional, Amersham, UK.) at 37C for one hour with shaking. After multiple washes the targets were resuspended in RPMI 1640 at a density of 5 X 104 ml.- 1 Labelled targets (5 x 103 well- 1) were dispensed into the wells of 96-well U-bottomed microtitre plates (Sterilin Ltd., Feltham, England) and then graded numbers of effector cells were added to give effector-to target ratios (E:T) ranging from 40:1 to 1.25:1 in a total of 200 µl. of medium. After a four-hour incubation at 37C and cen trifugation at 100 g for five minutes, 100 µl. of supernatant was removed from each well and the radioactive content determined using a liquid scintillation counter (Canberra Packard Ltd.). All assays were performed in triplicate and the data were calculated from E:T of 40:1 in order to allow comparison between lines of differing susceptibility and expressed as the percentage specific cytotoxicity, calculated as below;
Monoclonal Antibody
Specificity
Source
CD3
T lymphocytes
BD
CD4
Helper T cells
SAPU
CD8
Cytotoxic T cells
SAPU
CD16
NK cells
BD
IL-2R
Receptor for interleukin-2
Amersham
HNK-1
Fe receptor on NK cells
BD
BD - Beclon Dickenson Ltd. (Cowley, UK) SAPU - Scottish Antibody Production Unit (Carluke, UK) Am-Amersham International (UK)
1
1
1
1
1
1
51
%Specific Cytotoxicity cpm exp release - cpm spontaneous release cpm total release - cpm spontaneous release Source of monoctonal antibodies. For details of antibody spec ificity and source see table 1. Monoclonal antibodies which were not directly conjugated to a fluorochrome were detected using a goat anti-mouse-fluorescein isothiocyanate (FITC) con jugate (Sigma Chemical Co., USA). Fluorescence staining and flow cytometric analysis. Two col our immunofluorescence was performed on effector cells
stained with monoclonal antibodies conjugated to either FITC or phycoerythrin. For each analysis 5 x 105 cells were washed with washing buffer (PBS, 1% FCS, 0.01% sodium azide) and incubated at 4C for 30 minutes with optimal concentrations of primary monoclonal antibody. Following two washes, the bind ing of unconjugated monoclonal antibodies was detected using a goat anti-mouse-FITC conjugate for a further 30 minutes at 4C. The cells were then blocked with a 10% (v/v) solution of normal mouse serum for 30 minutes and, after washing, incu bated with a second monoclonal antibody conjugated to phy coerythrin for a further 30 minutes. The cells were washed twice more and resuspended in 1% formaldehyde prior to analysis using a flow micro-fluorometer (Coulter, EPICS-C, USA). Non-viable cells were gated out of the window and at least 10,000 events were accumulated by using logarithmic amplification of fluorescence intensity. Statistical analysis. All experiments were performed a mini mum of five times in triplicate. Statistical analysis was per formed using the non-para-metric "Mann-Whitney U" test. The data were analysed using Statview 512 software on an Apple Macintosh Hex computer. RESULTS
Dose dependent kinetics of LAK cell induction. Although previous studies have demonstrated that IL-2 is capable of generating from PBMC LAK cells directed against a variety of tumour cells, little work has been undertaken on non-renal urological malignancies and in particular TCC of the bladder. In order to evaluate the ability of LAK cells to mediate cytolytic activity against bladder cancer, normal PBL were cultured with increasing doses of IL-2 for up to 10 days. The results clearly demonstrate that IL-2 can induce PBL to exhibit LAK activity against a panel of bladder cancer cell lines (figure 1). The induction of LAK activity was found to be dependent upon the dose of IL-2 and the duration of the activation period. As little as 10 Uml.- rIL-2 was able to induce LAK activity against all four cell lines after only three days activation. However, optimal levels of LAK activity were achieved after six days incubation with 1,000 Uml.- IL-2. Any further increase in the dose of IL-2 or in the duration of the activation period failed to result in a corresponding increase in lytic potential (data not shown). The four cell lines investigated were not killed with equal efficiency by the same population of effector cells. The grade 2 tumour, RT112, was least susceptible to LAK mediated cytol ysis (mean± standard error, 22.9%± 2.5), RT4 (Gl) (49.2%± 6.0) and MGH-Ul (G3) (49.1% ± 3.9) were more susceptible, and finally EJ18 (G3) was most susceptible (62.3%± 3.9). Influence of PBMC donor on LAK killing. The cytotoxic potential of LAK cells derived from five donors was simulta neously assessed against the RT112 and MGH-Ul cell lines. The results, shown in figure 2, demonstrate slight differences 1
1
209
LAK ACTIVITY AGAINST BLADDER TUMOUR CELLS IN VITRO
..
�
:�
��
u (J
,.:: ·i:;
.,...
.."....
00
i:,.
·�1
Ill
� l!lil �
RT4
,.
100 ulL-2/Mi 1000 aBL-2/ml
�_J
20
•
.."'
100
� 40 hours
i....
[:2l
"
