Cancer Immunol Immunother (1992) 35:381-387

ancer mmunolggy mmunothPapy © Springer-Verlag 1992

Interferon y but not tumor necrosis factor decreases susceptibility of human renal cell cancer cell lines to lymphokine-activated killer cells Yoshihiko Tomita I , Hisami Watanabe 2, Hisashi Kobayashi 3, Tsutomu Nishiyama 1, Shoji Tsuji 3, Michio Fujiwara 2, and Shotaro Sato 1 1 Department of Urology, 2 Department of Immunology, Niigata University School of Medicine, and 3 Department of Neurology, Brain Research Institute, Niigata University Niigata, Japan Received 4 December 1991/Accepted 19 June 1992

Summary. Human renal cell cancer (RCC) cell lines, ACHN and KRC/Y, with or without exposure to cytokines, were examined for their susceptibility to lymphokine-activated killer (LAK) cells. Flow-cytometric analysis demonstrated constitutional expression of class I antigen on both cell lines, which was enhanced by interferon o~ (IFN(~), IFNy and tumor necrosis factor ¢x (TNFo0. A 4-h 51Cr-release cytotoxicity assay demonstrated that pretreatment of both cell lines with IFNy or IFN(z, but not with TNFo~, decreased their susceptibility to LAK cells. IFNy also decreased susceptibility to natural killer cells in a 16-h 51Cr-release cytotoxicity assay. IFNy treatment decreased the susceptibility of ACHN cells in a dose-dependent manner. "Cold"-target competition assay clearly showed that IFNy- but not TNF~-pretreated cells compete less effectively than do untreated target cells. Pretreatment with IFN 7, however, increased expression of intercellular adhesion molecule-1 (ICAM-1) to a degree comparable to that with TNF(z. Northern blot analyses using a 520-base-pair ICAM-1 cDNA as a probe demonstrated that more 3.3-kb mRNA is expressed in IFNy- and TNFc~-pretreated cells. These results suggest that IFNy-treated RCC cell lines may reduce their ability to be recognized by LAK cells, and that IFN-induced protection of RCC cell lines against LAK cells may depend upon a mechanism independent of the expression of class I antigens or ICAM-1 on tumor cells. Key words: MHC class I molecules - ICAM-1 - LAK cells

Introduction

Recently, several types of cytokines: interferons (IFN), interleukin-2 (IL-2) and tumor necrosis factor o~ (TNFc~),

Correspondence to: Department of Immunology, Niigata University School of Medicine, Asahimachi- 1, Niigata 951, Japan

or combinations of these, have been used for treatment of malignant tumors both clinically [1, 9, 22] and in murine systems [4, 25]. In these studies, antitumor effects of the above agents through activation of host immune cells, as well as a direct antitumor effect, were demonstrated [4]. Moreover, adoptive immunotherapy using lymphokine-activated killer (LAK) cells has been performed clinically since Rosenberg et al. first reported its effectiveness against malignant tumors [23]. Besides the usefulness of cytokines for treatment of cancer, it has been shown that pretreatment of tumor cells with IFN decrease their susceptibility to both LAK cells [6, 19, 32] and natural killer (NK) cells [10, 13, 31]. In most of these studies, up-regulation of major histocompatibility complex (MHC) class I antigen expression in tumor cells was closely related to the decrease in susceptibility to LAK cells. Moreover, concentration of IFN which can induce protection in such studies, can be achieved by their administration to patients with malignant tumor [33]. Therefore, clinically administered cytokines might influence the susceptibility of tumor cells to LAK and NK cells. Furthermore, it has been reported that IFN 7, produced by tumor-infiltrating lymphocytes, may induce phenotypic change in tumor cells [3], suggesting that this cytokine might induce protection against LAK cells. Renal cell cancer (RCC) is notable in that it shows a highest number of cases of spontaneous regression than any other malignant tumor [7] and has a relatively marked response to immunotherapy with cytokines [14] or LAK cells [23]. In view of these features, progression of RCC is considered to be correlated with the immune cells of the host. Although IFN-induced protection against LAK cells has been observed in many types of tumor in relation to an increase of class ! antigen expression, there appear to be no reports on RCC, for which adoptive immunotherapy is thought to be favorably indicated. Furthermore, in our previous study, it was shown that class I antigen expression is preserved to a greater extent in RCC cells than in other malignant tumors, and that this is correlated with the degree of mononuclear cell infiltration [30]. Therefore, it is of interest whether the various types of cytokine used clini-

382

cally influence the susceptibility of RCC to LAK cells in relation with MHC class I antigen expression. In the present study, it was showed that exposure of RCC to IFN~, IFNy and TNF~ increased the expression of MHC class I antigen, whereas a marked decrease in the susceptibility to LAK cells was observed following pretreatment with IFNy, but not with TNFo~.

Materials and methods Reagents and cell lines. Monoclonal antibodies (mAb) used in this study are as follows: 84I-I10 against ICAM-1 (CD54) (Immunotech, S. A., Marseille, Cedex 9, France), W6/32 (IgG2a) against a monomorphic determinant on the heavy chain of MHC class I antigens associated with [~-2 microglobulin (Dako Japan Co., Kyoto, Japan), L243 (IgG2a) against a monomorphic determinant of HLA-DR (Becton Dickinson, Mountain View, Calif.). Recombinant IL-2 (rIL-2) and rlFNa were supplied by Takeda Pharmaceuticals, Osaka, Japan. rIL-2 had a specific activity of 3.5 x 104 U/mg protein as assayed on IL-2-dependent murine NKC3 cells [12]. rIFNy and natural (n) TNF~ were supplied by Shionogi Pharmaceuticals, Osaka, Japan, and Mochida Pharmaceuticals, Tokyo, Japan, respectively. The RCC cell line KRC/Y [35] was kindly provided by Dr. M. Kojiro, 1st Department of Pathology, Kurume University School of Medicine, Kurume, Japan. The ACHN cell line (ATCC no. CRL-1611) was kindly provided by Sumitomo Pharmaceuticals, Osaka, Japan. These cell lines grew in a monolayer and were subcultured after 5 rain of 0.125 % trypsin/0.01% EDTA treatment.

Cytokine treatment and flow-cytometric analysis. ACHN and KRC/Y were initially treated with a fixed dose of rlFNc~ (2500 U/ml), rlL-2 (2000 U/ml), rIFN'/and nTNF(x (both 500 U/ml) for 48 h. Both cell lines were then subjected to flow-cytometric analysis and cytotoxicity assay, and the results compared with those for untreated cells. Furthermore, dose/response experiments were performed on ACHN cells. For flowcytometric analysis, ceils were stained by an indirect immunofluorescence method as described before [34]. Briefly, tumor cells (1 x 106) were reacted for 30 rain at 4°C with mAb at a final dilution of 1:10 in phosphate-buffered saline supplemented with 2% fetal calf serum and 0.02% sodium azide. After two washes by centrifugation, the cells were incubated with fluorescein-isothiocyanatl-conjugated goat anti-mouse Ig (Tago Inc., Burlingame, Calif.) for 30 rain at 4°C at a final dilution of 1 : 30. Subsequently the cells were washed three times, suspended at a concentration of 1 x 106 cells/ml and analyzed by flow cytometry using FACScan (Becton Dickinson, Mountain View, Calif.). For examination of nonspecific binding, tumor cells were treated with normal mouse Ig instead of monoclonal antibodies. On a histogram of the negative control, a marker was set as positive cells were lower than 1%. Tumor cells that represented a higher fluorescence intensity than this marker were then counted as positive cells. When the examined molecule was constitutionally expressed, the mean fluorescence intensity of positive cells was compared to determine the influence of cytokine treatment. Generation of lymphokine-activated killer cells. After informed consent had been obtained, heparinized blood was drawn form normal volunteers and from patients with RCC (who were admitted to the Department of Urology, Niigata University School of Medicine). These patients had not received any type of therapy. Peripheral blood mononuclear cells (PBMC) were separated using Ficoll-Paque density gradients (Pharmacia LKB Biotechnology Inc., Piscataway, N. J.). PBMC were used for assaying natural killer (NK) activity against tumor. For LAK induction, PBMC were suspended at a concentration of 2 × 106 cells/ml in RPMI1640 medium containing 10% fetal calf serum and rlL-2 was added at a concentration 2000 U/ml. PBMC were cultured in 3013 (Falcon Labware, Oxnard, Calif.) 25-cm 2 tissue-culture flasks for 5 - 7 days at 37 ° C in a 5% CO2 atmosphere.

Cytotoxicity assay. Tumor target cells were labeled for 1 h by addition 3.7 MBq sodium [sICr]chromate (New England Nuclear, Boston, Mass.) after incubation with or without cytokines for 48 h. The ceils were washed four times by ceutrifugation and suspended at a concentration of 1 x 105 cells/ml. Samples of 100 gl of these target cells and (2.520) × 105 effector cells were added to each well of 96-well round-bottomed plates (Costar, Cambridge, Mass.). Spontaneous release of 5lCr was determined by incubation of target cells in medium alone and maximum release by addition of 100 gl 1% Triton X-100. Plates were incubated for 4 h at 37°C and the amount of 51Cr released into a 100-gl sample of the supernatant from each well was determined. In addition, for the assay of susceptibility to natural killer activity, from 1 x 106 to 4 x 106 effector cells were added and the lysis of target cells was measured after 16 h incubation. The percentage of specific cell lysis was calculated using the formula: experimental release - spontaneous release maximum release - spontaneous release All experiments were done in triplicate, and spontaneous 51Cr release from the various targets used was up to 15% of the maximum release. Furthermore, determination of kinetics of LAK-mediated lysis was performed. ACHN cells were labeled with 51Cr as described above, and added into 96-well plates at 1 × 104 cells/well. LAK cells were added at 1 × 105 cells/well resulting in an effector-to-target cell ratio of 10 : 1. Identical sets of wells for determining spontaneous release, maximum release, and specific cytotoxicity at various time points were set up at time 0. The plates were incubated at 37 ° C for 1, 2, 4, 6 and 8 h and the percentage specific lysis was determined. Specific lysis (%) =

"Cold"-target competition assay. Untreated tumor target ceils were labeled with 51Cr as described above and dispensed into 96-well plates at 5 × 103 cells/well in a 50-gl volume, Unlabeled target cells, IFN~,-treated (500 U/ml, 48 h), TNFc~-treated (500 Uhnl), 48 h), and untreated, were added to the wells at unlabeled/labeled target cell ratios ranging from 0/1 to 40/1. LAK cells were then added at 5 × 103 cells/well in 50 gl and lysis of labeled target cells was determined after a 4-h incubation as described. RNA extraction and Northern blot analysis of lCAM-1 on ACHN cells. A 520-base-pair ICAM- 1 cDNA probe was kindly provided by Dr. K. Imai (Sapporo Medical School, Sapporo, Japan) [20]. A ~-actin cDNA probe, which was kindly provided by Dr. H. Hamada (Tokyo University, Faculty of Medicine, Tokyo, Japan), was used as a control probe [ 11]. ACHN cells were cultured in 3375 (Costar, Calif.) 76-cm 2 tissue-culture flasks in RPMI-1640 medium containing 10% fetal calf serum with or without exposure to IFN 7 or TNFa for 48 h. These cells were harvested by cell scraper and washed twice with phosphate-buffered saline, and were extracted using a guanidinium isothiocyanate procedure to yield a total RNA preparation [5]. Samples containing 20 gg total RNA were size-fractionated on 1% agarose/formaldehyde gel and transferred to a nylon membrane (Gene Screen, NEN). Hybridization was performed at 65 ° C in 7% polyethylene glycol and 10% sodium dodecyl sulfate [28] using a multi-primed 32P-labeled probe (Amersham, Multiprime DNA labeling system, RPN 1600). Filters were washed finally in 0.2 x standard saline citrate (SSC; 1 x SSC = 150 mM NaC1, 15 mM sodium citrate) for 30 rain at 65 ° C, and autoradiographed using an intensifying screen at -80 ° C.

Proliferation assay. Effect of IFNy on proliferation of ACHN cells was assayed in microtiter wells (96-well flat-bottomed plates, Nunc, Roskilde, Denmark) by uptake of [3H]thymidine ([3H]dT). ACHN cells (1 x 104/well) were added to 96-well microculture plates and cultured for 24 h. Then, cytokines were added and cultured for48 h. [3H]dT (0.5 #Ci) was pulsed to each well and cells were harvested after 20 h of culture. [3H]dT incorporation was counted by a liquid scintillation counter.

383 Table 1. Effect of cytokines on HLA class I antigen expression on renal

A

cell cancer (RCC) cell linesa Cell lines

B

Positive cells (%) aftertreatmentwith: -

rIFNt~

rIFN7

I

nTNFct rIL-2

Class I antigens ACHN 99.6 (80) b 99.5 (172) 99.1 (129) 99.9 (173) 99.3 (62) KRC/Y 98.8 (109) 99.4 (143) 99.9 (274) 99.9 (207) 100.0 (100) HLA-DR ACHN KRC/Y

...... i~ ...... i~i ::"~-i'a" ...... i'~ ~ 0.4 11.1

0.3 11.4

12.4 58.2

1.4 13.2

eND cND

M H C class I antigens were expressed on nearly all of the A C H N and KRC/Y cells (Table 1). Both tumor cell lines were initially treated with a fixed dose of each of four different cytokines. As presented in Table 1, rIFNo~, rlFNy and nTNFc~ increased the expression of class I antigen on both cell lines. Furthermore, dose/response experiments were performed to examine the effects of rIFNo~ and rIFNy. Concentrations as low as 250 U/ml IFNo~ and 50 U/ml IFNy enhanced the expression of class I antigens, and maximal effects were shown at concentrations of 1000 U/ml IFN(z and 400 U/ml IFNy (data not shown), rIL-2 did not increase class I antigen expression. In addition, the effects of IFNo~, -y and TNFo~ on expression of H L A - D R were examined. IFN-y but not -o~ and TNFo~ increased H L A - D R expression on both cell lines (Table 1). Dose/response experiments on H L A - D R induction on A C H N cells demonstrated an increase of H L A - D R when cells were pretreated with IFN 7 at higher concentration than 50 U/ml for 48 h (Fig. 1).

Effect of cytokines on susceptibility to LAK cells Among the cytokines tested, exposure to WNy markedly decreased the susceptibility of A C H N to L A K cells and moderately decreased the susceptibility of KRC/Y; I F N ~ had a moderate effect on both cell lines (Table 2, experiments 1 and 2). Treatment with TNF(z, however, did not markedly influence the susceptibility despite the increase of class I antigens on treated tumor cells. The effects of IFNy treatment were also confirmed in cells from three other normal donors (Table 2, experiment 3), and the results were consistent. Statistical analysis of three experiments revealed a significant difference between untreated and IFNT-treated cells at each effector: target ratio examined: for A C H N cells 5:1, P