Acta path. microbiol. scand. Sect. C, 89: 235-239, 1981

CELLULAR AND HUMORAL IN VITRO CYTOTOXICITY AGAINST AUTOLOGOUS BLADDER TUMOR CELLS

IN HUMANS

Dqferences in Autologous Cytotoxiciry against Non-invasive and Invasive Transitional- cell Tumors of the Urinary Bladder

FLEMMING JACOBSEN

University Institute of Pathology, Aarhus Amtssygehus, DK-8000 Aarhus C. Denmark

Jacobsen, F. Cellular and humoral in vifro cytotoxicity against autologous bladder tumor cells in humans. Differences in autologous cytotoxicity against non-invasive and invasive transitional-cell tumors of the urinary bladder. Acta path. microbiol. scand. Sect. C, 89: 235-239, I98 I .

Lymphocyte-mediated cytotoxicity (LMC), antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) were measured in a slCr assay against autologous tumor cells from 7 patients with non-invasive and 9 patients with invasive transitional-cell tumors of the urinary bladder (TCC). Cytotoxicity against the invasive tumor cells were demonstrated in CDC. Heat- inactivation of sera lead to complete loss of cytotoxicity while addition of allogenic serum restored cytotoxicity. The cytotoxicity of allogenic sera from controls against inwasive tumor targets showed no differences when compared with autologous sera. No or very weak cytotoxicity was found against non- invasive tumor targets in autologous or allogenic assays. LMC and ADCC showed weak or no reactivities. Intensive wash or trypsinization of effector cells did not affect the cytotoxicity in LMC. The results indicate the occurrence of complement-dependent antibodies directed against target cells from invasive tumors of the urinary bladder while no cytotoxic responses were detectable when the target cells originated from non-invasive tumors.

Key words: Cellular cytotoxicity; humoral cytotoxicity; bladder tumors.

F. Jacobsen. University Institute of Pathology, Aarhus Amtssygehus, DK-8000 Aarhus C , Denmark.

Received 4.ix.80 Accepted 1 i . 8 I Conflicting results exist concerning the role of

cell-mediated and humoral immune response in human cancer (Bean et al. 1975, Perlmann et al. 1977). The in vitro cytotoxicity of blood lymphocy- tes from cancer patients against established cell lines or short-term cultures of cancer cells have been examined in autologous and allogenic assays with appropriate controls of target and effector cells (Baldwin & Embleton 1977). In the interpretation of the results discrepancies have - apart from technical variations - been related to ))natural cytotoxicitycc displayed by effector cells from controls (Troye et al. 1977, Herberman et al. 1979, Santoli & Koprowski 1979).

Lymphocytes from patients with transitional-cell carcinomas of the urinary bladder (TCC) were reported to express a disease-related cytotoxicity superimposed on a natural cytotoxicity (Vilien & Wolf 1978, Troye et al. 1980 a). Further studies have indicated that the cytotoxic effector cells are lymphocytes expressing Fc-receptors for IgG sug- gesting an antibody-mediated cytotoxicity Vape et al. 1977). When cytotoxic lymphocytes expressing Fc-receptors for immunoglobulins are armed or blocked with antibodies or immune complexes, removal of antibodies by intensive wash or trypsini- zation might affect the cytotoxicity. From studies of the antibody-dependent cell-mediated cytotoxicity (ADCC) of patients with TCC, Troye et al. ( 1980 b)

235

suggested that these patients may develop a disease- related humoral immune response. Studies of complement-dependent cytotoxicity (CDC) indicated a higher incidence of circulating tumor-directed antibodies in cancer patients when compared to control donors (Lewis et al. 1969, Nairn et al. 197 I , Bodurtha et al. 1975). Bubenik et al. (1 970) reported that sera from 8 of 13 patients with bladder carcinoma had complement-dependent antibodies cytotoxic to cultured autologous and allogenic bladder tumor cells.

The purpose of the present study has been to investigate and compare autologous and allogenic cellular and humoral cytotoxicity against non- cultivated tumor cells from patients with non- invasive and invasive tumors of the urinary bladder.

MATERIAL AND METHODS

Patients Sixteen patients ( I I males and 5 females, aged 34-85

years, mean 66 years) with tumors of the urinary bladder (TCC) were included in this study. None had received treatment for their disease. Twelve other patients were excluded for following reasons: three specimens did not contain tumor tissue, 4 specimens revealed scarcely any target cells for testing, in 4 cases the spontaneous release exceeded 50% (see below) and 1 specimen was excluded since the patient had received local chemotheraphy.

Clinical staging of the tumors was done according to U.I.C.C. (Geneva 1979) and histological grading of the tumors according to Bergkvisf et al. ( 1 965). Stages and malignancy grades are given in Table 1 .

Controls Sera and lymphocytes from 16 healthy donors (HD)

(aged 2 5 - 4 0 years) were used as controls. The donors were unmatched for sex. In CDC, allogenic sera from I I patients (6 males and 5 females, aged 56-77 years, mean 73.2 years) with 6 invasive and 5 non-invasive tumors of the urinary bladder were used as clinical controls (CC). None had received treatment for their disease.

Target Cells The tumor target cells were isolated from fresh tumor

biopsies. The biopsy specimens were collected in RPMI-I 640 medium supplemented with 10% (v/v) heat-inactivated fetal calf serum. I mM glutamine, 50 IU penicillin/ml. 50 mcgs streptomycin/ml and 10 mM Hepes buffer (pH 7.38) (Microbiological Associates, Walkersville. Maryland 2 1793). The urothelial tumor cells were removed under a stereomicroscope by scraping with a lancet. The target cells were incubated for 30 minutes at 37 OC in a solution containing 0.25 % trypsin (Difco Laboratories. Detroit, Michigan) and 0.02 % EDTA (BDH Chemicals Ltd.. Pool, England). The enzymatic process was stopped by dilution with medium followed by filtration through two layers of 28-mesh gauze. Gradient centrifugation was performed in Lym-

236

foprep (Nyegaard & Co A/S, Oslo, Norway (relative density 1,077)) at 600 g for 20 minutes. From the interface a single cell suspension of tumor cells was harvested.

Effector Cells Blood samples from patients were taken simulta-

neously with collection of tumor specimens. In order to isolate mononuclear cells blood samples were obtained in EDTA (2 drops 30 % EDTA per ml blood). Lymphocy- tes were isolated according to the method described by Boyum ( I 968).

In certain experiments intensive wash of the effector cells by centrifugation at 600 g for 5 minutes, removal of supernatant and resuspension in Hanks’ solution was performed ten times before final resuspension in medium.

In other experiments the effector cells were incubated in 0.25 % trypsin solution for 30 minutes at 37 OC. The cells were then washed twice and adjusted to the desired cell concentration.

Sera Sera from controls were stored for a few days at -

20 OC. Heat-inactivation of sera was performed at 56 OC for 30 minutes. Freshly prepared serum from one healthy donor was used as complement source.

Cytotoxicity Assays The LMC cytotoxicity assay was performed immedia-

tely after isolation of target and effector cells as a 5lchromium release assay. The target cells were washed once and incubated with 50 pCi Na2 5lCr 0 4 (Radiochemical Centre, Amersham, England (specific activity 100-400 mCi/mg)) per 5 x 105 target cells for 60 minutes at 37 OC. After two washes in medium the target cell suspension was adjusted to contain 5 x 104 cells per ml. The assay was carried out in round-bottom I0 x 70 mm test tubes (Nunc, Denmark). Every test was done in duplicates each tube containing 5 x 103 target cells. Effector cells were tested in an effector cell/target cell ratio of 50/1.

ADCC was performed with donor lymphocytes and autologous inactivated and undiluted sera. CDC was performed with undiluted sera.

The incubation period was 6 hours at 37 OC. The spontaneous isotope release was estimated in medium alone. Final assay volumes were 0.2 ml or 0.3 ml. At the end of the incubation 0.8 ml or 0.7 ml of medium, respectively, was added to each test tube, and the cells were resuspended. Half of the supernatant was removed after centrifugation for 5 minutes at 600 g and counted in a gamma counter. The remaining supernatant and the cell pellet (residue) were counted. The background activity was subtracted from the results. The percentage of isotope release (corr. % ) was calculated as follows:

supernatant x 2 supernatant + residue

Test x 100 minus

supernatant x 2 supernatant + residue

spontaneous x 100

Experiments in which the spontaneous isotope release exceeded 50% were excluded. In the present study spontaneous release ranged between 2 1 and 49 per cent. Other authors have accepted a similar high background activity in assays using tumor cells separated from biopsy specimens (Vose et al. 1977, Vanky & Argov 1980). There was no significant difference in the spontaneous release between non-invasive TCC (mean * SD: 36.4% * 12.1 %) and invasive TCC (mean* SD: 29.8 % * 8.7 % ). The standard deviation for differences within duplicates did not exceed 3 per cent.

Statislics Statistical analysis of the results was performed by use

of the Wilcoxon's rank sum test for paired observations and the Mann-Whitney's rank sum test for unpaired observations. Differences in cytotoxicity were considered to be of significance at p values below 0.05.

% Isotope Release ( corr. 5 I I 20 I

!

RESULTS

The relationships between the clinical stages and malignancy grades of 7 non-invasive TCC and 9 invasive TCC and the cytotoxicity of autologous lymphocytes and autologous sera are presented in Table 1 . The results of the cytotoxicity tests are corrected for the spontaneous isotope release. The comparisons of autologous cytotoxicity exerted by

N.S.

Lymphocytes

p ' 0.01

Sera

Fig. I. Comparison of the cytotoxicity induced by autologous lymphocytes in LMC and autologous sera in CDC against non-invasive and invasive bladder tumor cells. Horizontal lines: mean cytotoxicity Corr. % : corrected for spontaneous isotope release N.S.: no significant difference 0 : non-invasive bladder tumor cells 0 : invasive bladder tumor cells

TABLE 1. Cytotoxicity oJ Autologous Lymphocytes in LMC and Autologous Sera in CDC in Relation to Stage and Grade of 7 Non-invasive and 9 Invasive Tumors of the Urinary Bladder

Patients Inactivated

sera and complement

Inactivated sera

Lymphocytes Sera 50/ la Stage Grade

H.M. Ta I 6b I - I - I M.H. Ta 1 3 0 0 I C.S. Ta I1 6 0 -I - I E.N. Ta I1 3 3 2 4 H.S. Ta 11 0 3 0 I A.H. Ta I1 -3 I 1 -3 I.L. Ta I1 -1 -5 -7 -6

S.O. S.A. F.N. H.S. M.L. J.K. A.L. E.L. N.J.

T- 1 T- 1 T- 1 T-2 T-2 T-2 T-3 T-3 T-4

I1 I11 I11 I1 I11 111 I11 IV IV

0 -4 - I

1 -3 -1 12 2 2

20 6 5 3

19 6

19 10 24

-2 -5 -2 -1 -2 -1

2 3 0

13 I 3 1

16 3 3

12 23

a effector cell/target cell ratio corrected for spontaneous isotope release

2 3 1

TABLE 2 . Comparison of the Mean Cv1otoxicit.v in CDC Induced by Autologous Sera, Sera from Healthy Donors (HD) and Sera from Clinical Controls (CO

Inactivated Target cells Autologous sera autologous sera

and complement HD Inactivated

autologous sera cc

Non-invasive 7a 0.4%b(2.7) 7 -0.9% (2.9) 7 0.7% (3.2) 7 1.3% (3.4) 10 0.8% (3.4)

Invasive 9 12 .5% (8.0) 9 -0.9% (2.4) 9 8.3% (7.9) 8 15.0% (13.2) 8 10.9% (8.2)

p(O.01 ns p<O.OI p<O.Ol p<O.OI

a

ns no significance b

( ) standard deviarion

number of persons included in calculation

corrected for spontaneous isotope release

lymphocytes and sera against non-invasive and invasive tumor targets are shown in Fig. 1 . In LMC no cytotoxicity was demonstrable. In CDC signifi- cant differences in cytotoxicity between non- invasive and invasive tumor cells were seen (p<O.OI). indicating a cytotoxic effect of autolo- gous sera against invasive tumor cells.

No cytotoxicity was excerted by either autologous or allogenic lymphocytes. In an attempt to remove arming or blocking antibodies on lymphocytes, intensive wash or trypsinization of lymphocytes were performed and the cytotoxicity against autolo- gous tumor cells was measured. Intensive wash or trypsinization did not affect the activity of the effector cells. The cytotoxicity induced in ADCC is independent of complement. ADCC was determi- ned using donor lymphocytes and autologous inactivated sera. The results did not reflect any cytotoxicity against the tumor cells.

Table 2 shows the cytotoxicity excerted by autologous and allogenic sera. Significant differen- ces in cytotoxicity against non-invasive and invasive target cells in autologous and allogenic combina- tions were found (p<O.OI). Complete loss of cytoroxicity was seen after heat-inactivation of autologous sera. Addition of a fresh allogenic serum to inactivated autologous sera restored the cytotoxi- city. h considerable scatter in the cytotoxic reactivi- ties against invasive target cells was observed. However. the parterns of reactivities were equal for the tested autologous and allogenic sera.

DISCUSSION

Complement-dependent cytotoxicity was demon- strated against autologous tumor cells from histolo- gically verified invasive bladder tumors. However, allogenic sera from healthy donors (HD) and

238

patients with non-invasive and invasive bladder tumors (CC) revealed the same mean cytotoxicity against invasive tumor cells as the autologous sera. Thus, the present results do not substantiate a disease-related humoral immune response superim- posed on a ))natural(( immunity as suggested by Troye et al. (1 980 b).

In LMC no significant differences were noted in autologous or allogenic combinations against non- invasive and invasive tumor targets (Jacobsm 198 1 a). The slight reactivities in LMC could not be enhanced by intensive wash or trypsinizatidn of effector cells and accordingly the results do not demonstrate cytophilically carried antibodies with either arming or blocking function. The cytotoxic effector cells seem to belong to the non-T cell fraction of lymphocytes (Jacobsen 198 I b).

In the present study autologous heat-inactivated sera and donor lymphocytes did not induce ADCC as suggested by other studies using cultured target cells ( T r o y et al. 1980 b). The results of the present study suggest the existence of natural antibodies against invasive bladder tumor cells. The comple- ment-dependent serum-mediated cytotoxicity was strongly dependent on the nature of the target cell. This finding suggests differences in surface antigens between non-invasive and invasive bladder tumor cells. Unmasking of antigens on the malignant tumor cell may be a result of the initial mechanic and enzymatic disintegration of the tumor tissue. The marked variation in cytotoxicity against inva- sive tumor targets may be due to differences in the degree of unmasking or in antigenic armamentation of the invasive target cell population.

An explanation why sera from healthy donors and clinical control patients reacted at equal levels in CDC could be that the tumor cells were coated with antibodies in advance (Thunold et al . 1973. Izsak rt al. 1974). Should this be the case, the cytotoxic

event would only require a complement source. Differences in cytotoxicity may be explained by differences in the density of complement-fixing immunoglobulins on the surfaces of the tumor cells.

In summary, the present study indicates the existence of complement-dependent antibodies di- rected against target cells from invasive tumors of the urinary bladder. The cytotoxic response seems strongly dependent on the type of target cell used. Further characterization of the responsible antibo- dies, the surface antigens of the tumor cells and the proposed immunoglobulin coating of tumor cells are needed.

REFERENCES

Baldwin, R . W . & Embleton, M . J.: Assessment of cell- mediated immunity to human tumorassociated anti- gens. Int. Rew. Exp. Path. 1 7 : 49-95, 1977.

Bean, M. A . , Bloom, B. R . , Herberman, R . B.. Old, L . J., Oettgen, H . F., Klein. G . & Terry, W . D.: Cell- mediated cytotoxicity for bladder carcinoma: Evalua- tion of a workshop. Cancer Res. 35: 2902-2913, 1975.

Bergkvist, A , , Ljungquist, A . & Moberger, C.: Classifica- tion of bladder tumors based on the cellular pattern. Acta Chir. Scand. 130: 371-378, 1965.

Bodurrha. A . J . , Chee, D. 0.. Laucius, J. F., Mastrangelo, M. J. & Prehn, R . T.: Clinical and immunological significance of human melanoma cytotoxic antibody. Cancer Res. 35: 189-193, 1975.

Boyum, A , : Isolation of mononuclear cells and granulo- cytes from human blood. Scand. J. Clin. Lab. Invest. 21, SUPPI. 97, 77-89, 1968.

Bubenik, J., Perlmann. P., Helmstein, K . & Moberger, G.: Cellular and humoral responses to human urinary bladder carcinomas. Int. J. Cancer 5: 310-319, 1970.

Herberman, R . B. . Djeu. J. Y . . Kay, H . D. . Ortaldo. J. R . . Riccardi. C . , Bonnard, C. D . . Holden, H . T. . Fagnani, R . , Santoni, A . & Puccetti, P.: Natural killer cells: Characteristics and regulation of actiyity Immunological Rev. 44: 43-70, 1979.

Izsak, F. C . , Brenner, H . J . . Landes, E. , Ran, M . & Witz. I . P.: Correlation between clinico-pathological featu- res of malignant tumors and cell surface immunoglo- bulins. Isr. J. Med. Sci. 10: 642-646, 1974.

Jacobsen, F: Lymphocyte cytotoxicity against autologous bladder tumor cells in humans. Acta path. microbiol. scand. Sect. C, 89: 175-180, 1981 a.

Jacobsen, F.: In v i m cytotoxicity of lymphocyte subpopulations against autologous human bladder

tumor cells. Acta path. microbiol. scand. Sect. C 89:

Lewis, M . G. , Ikonopisov. R . L . , Nairn, R . C . , Phillips, T . M . , Hamilton Fairley. C.. Bodenham. D . C . & Alexander, P.: Tumour-specific antibodies in human malignant melanoma and their relationship to the extent of the disease. Brit. Med. J . 3: 547-552. 1969.

Nairn. R . C., Nind, A . P. P. , Guli, E . P. C., Muller, H . K.. Rolland, J . M. & Minty, C . C . J.: Specific immune response in human skin carcinoma. Brit. Med. J. 4 :

Pape, C . R . , Troye, M . & Perlmann, P.: Characterization of cytolytic effector cells in peripheral blood of healthy individuals and cancer patients. 11. Cytotoxi- city to allogenic or autochthonous tumor cells in tissue culture. J. Immunol. 118: 1925-1939, 1977.

Perlmann, P., Troye. M. & Pape, G. R . : Cell-mediated immune reactions to human tumors. Cancer 40: 448-457, 1977.

Santoli, D. & Koprowski, H . : Mechanisms of activation of human natural killer cells against tumor and virus-infected cells. Immunological Rev. 44: 125- 163, 1979.

Thunold. S., Tonder, 0 & Larsen, 0.: Immunoglobulins in eluates of malignant human tumours. Acta path. microbiol. scand. Sect. A suppl. 236, 97-100, 1973.

Troye. M . , Perlmann, P.. Larsson. d., Blomgren, H . & Johansson, B.: Cellular cytotoxicity in vitro in transitional cell carcinoma of the human urinary bladder. 5 1 Cr-release assay. Int. J. Cancer 20: 188- 198, 1977.

Troye, M . , Vilien, M . , Pape, C. R . & Perlmann, P.: Cytotoxicity in vitro of blood lymphocytes from bladder cancer patients and controls to allogenic or autologous tumor cells derived from established cell lines or short-term cultures. Int. J. Cancer 25: 33- 43, 1980 a.

Troye, M. . Hansson, Y . , Paulie. S . , Perlmann. P. , Blomgren. H . & Johansson, B.: Lymphocyte-me- diated lysis of tumor cells in vitro (ADCC), induced by serum antibodies from patients with urinary bladder carcinoma or from controls. Int. J. Cancer

Union Internationale Contre le Cancer. TNM classifica- tion of malignant tumors. UICC. Geneva, 1979.

Vanky, F. & Argov, S . : Human tumor - lymphocyte interaction in vitro. VII. Blastogenesis and genera- tion of cytotoxicity against autologous tumor biopsy cells are inhibited by interferon. Int. J. Cancer 26:

Vose, B. M . , Vanky, F. & Klein, E . : Lymphocyte cytotoxicity against autologous tumor biopsy cells in humans. Int. J. Cancer 20: 5 12-5 19, 1977.

Vilien. M . & Wolf, H.: The specificity of the microcyto- toxicity assay for cell-mediated immunity in human bladder cancer. J. Urol. 119: 338-342, 1978.

241-246, 1981 b.

701-705, 1971.

25: 45-51, 1980 b.

405-41 I , 1980.

239