(CANCER RESEARCH 57. 4557-4563, October 15. 1W7|

Inieri eukin 18 Induces the Sequential Activation of Natural Killer Cells andCytotoxic T Lymphocytes to Protect Syngeneic Mice from Transplantationwith Meth A Sarcoma1

Mark J. Micallef,2 Tadao Tanimoto, Keizo Kohno, Masao Ikeda, and Masashi Kurimoto

Fujisaki Imtiiute, Hayashihara Biochemical Laboratories Inc., Fujisaki 675-1. Okayama 702. Japan

ABSTRACT

We have recently reported that interleukin 18 (IL-18) pretreatment

induces ¡mmunologically mediated antitumor effects in BALB/c miceinjected i.p. with syngeneic Meth A sarcoma. In this study, mice werepretreated with IL-18 before Meth A transplantation, and immunocom-petency in pretreated or untreated tumor-bearing mice (TBM) 3,9, and 15

days after transplantation was compared with that of normal mice. Onday 3, pretreated TBM mitogen-stimulated spleen cells produced significantly decreased levels of 11-2 and 11N-y during 24-h culture. In contrast,IL-10 and granulocyte macrophage colony-stimulating factor productions

were significantly enhanced in pretreated TBM cultures, and naturalkiller (NK) cell activity was also significantly augmented. Splenomegalywas also observed in the pretreated TBM on day 3, and the proliferatingcells were identified as asialo GM1+ cells by flow cytometry. Cytotoxic

activity of pretreated TBM spleen cells after a 5-day mixed lymphocyte-

tumor cell culture did not differ from that of untreated TBM and normalmice on day 3 but was significantly enhanced on days 9 and 15 comparedwith that observed in normal mice and untreated TBM. Concurrently, theproduction of IL-2 and of IL-10 recovered and decreased, respectively,and NK activity dropped to normal levels. The effects of IL-18 on cytokine

production and NK activity observed on day 3 treated TBM were alsoreproduced in normal mice. In conclusion, IL-18 seems to enhance the

generation of NK activity early after tumor transplantation and simultaneously induces an increase and a decrease in the production of IL-10 andIL-2, respectively. As NK activity subsides to normal levels and IL-10synthesis decreases, IL-2 synthesis is restored, and cytolytic cell activity issignificantly enhanced. These results provide new insight into the iiiiiiiu-nologically mediated antitumor effects of IL-18.

INTRODUCTION

A functioning immune system not only protects the host frominvasion by opportunistic foreign microorganisms but is also a systemof defense against syngeneic cells that have departed from cell growthcontrol networks that normally function to prevent uncontrolled cellgrowth and the evolution of neoplastic disease. Studies have shownthat malignant cells can develop mechanisms that prevent the induction of an effective antitumor immune response, leading to in vivoselection and the emergence of malignant cells, which are hardy andmore difficult for the immune system to eliminate (1,2). Models thatexplain how malignant cells escape from immunological surveillancenetworks and factors that stimulate an effective antitumor immuneresponse present new opportunities and target molecules for theimmunotherapy of malignant disease (3-5).

IL-18,3 which was previously known as IFN-y-inducing factor, is

an 18.3-kDa protein that induces the production of IFN-y by mouse

Received 4/29/97; accepted 8/29/97.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part by the Hayashibara Cancer Research Fellowship

Program (M. J. M.).2To whom requests for reprints should be addressed. Phone: 81-86-276-3141; Fax:

81-86-276-6885.' The abbreviations used are: IL. interleukin: NK. natural killer; TBM. tumor-bearing

mice; GM-CSF. granulocyte macrophage colony-stimulating factor; MLTC. mixed lymphocyte-tumor cell culture; MMC. mitomycin C; Th. T helper; Con A. concanavalin A.

cells in vivo and in vitro (6), augments mouse and human NK cellactivity (6, 7), enhances IL-2-mediated mouse and human T-cell

proliferation, and has a tendency to augment the production of Th type1 cytokine production by murine Th 1 clones and enriched human Tcells in vitro (8, 9). Murine NK cell clones established from the liverhave also been shown to exhibit enhanced Fas-mediated cytotoxiceffects when stimulated with murine IL-18 (10). Some of the immu

nological effects induced by this molecule are similar to those ofIL-12, a cytokine that is undergoing clinical trials in patients withmalignant disease; likewise, IL-18 was also considered a potential

agent for the induction of immunologically mediated antitumor responses in vivo.

We have recently reported that IL-18 exhibits significant antitumor

effects in BALB/c mice transplanted i.p. with the syngeneic sarcomaMeth A. especially when administered before tumor transplantation,and also induces immunological memory in the surviving mice (11).According to this treatment schedule, at least 90% of treated micesurvived long term after treatment. The in vivo antitumor effects ofpretreatment with IL-18 in the Meth A model were abrogated byeliminating NK cell activity after treatment with anti-asialo GM1

polyclonal antibody. Mice that survived injection with Meth A afterpretreatment with IL-18 spontaneously rejected a second transplanta

tion of the tumor cells about 3 months after the first tumor injection.Interestingly, the effector cells responsible for immunological memory in MLTCs during in vitro stimulation of mouse spleen cells withMMC-treated Meth A cells were found to be CD4+ T cells by

depletion with antibodies and rabbit complement. The immunologicalevents that occur between the induction of NK activity and theinduction of CD4+ memory T cells are the subject of the present

study.

MATERIALS AND METHODS

Experimental Animals. Eight-week-old female BALB/c mice were pur

chased from Charles River Japan. Inc. (Kanagawa. Japan) and used in ourexperiments 2-3 weeks after procurement. Mice were kept under conventionalconditions in a specific pathogen-free environment and provided with standard

feed and water ad libitum.Reagents. Murine IL-18 is a product of Hayashibara Biochemical Labora

tories (Okayama, Japan) and was produced and purified as described previously (6). Cell culture was performed in RPMI 1640 supplemented with 10%heat-inactivated fetal bovine serum (FBS; Whittaker. Walkersville, MD). pen

icillin (100 units/ml), and streptomycin (50 /ng/ml). referred from here onwardto as complete medium. Cells were incubated at 37°Cin a 5% CO, humidified

atmosphere. Meth A sarcoma grows aggressively in syngeneic BALB/c miceand was obtained from American Type Culture Collection (Rockville. MD).These cells were maintained in culture in vitro and occasionally passaged i.p.in syngeneic mice to ensure that the tumorigenic potential of the cells did notdiminish with culture.

In Vivo Treatment Protocols. Mice were injected i.p. with l /ig of IL-18or vehicle twice, 3 days and 6 h before being transplanted i.p. with 1 X IO5

Meth A cells. We have previously reported that this IL-18 pretrealment

schedule is very effective at preventing progressive tumor growth in vivo (11).Control normal mice were included in all our experiments. Untreated TBMseemed to be normal for the first week after i.p. transplantation of Meth A

4557

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTITUMOR KILLER CELL ACTIVATION BY IL-18

cells; however, from about day 10, the abdomens of the mice seemed toexpand, and a corresponding increase in body weight of the untreated mice wasrecorded. Mice pretreated with IL-18 did not accumulate malignant ascites.

The increase in body weight of the untreated TBM correlated well with anincrease in the accumulation of malignant ascites and is represented in Fig. 1as an indicator of when immunocompetency was examined in this study. Onday 15 after tumor transplantation, untreated TBM were still feeding well andmoved about freely. About 18 days after transplantation of the tumor cells,mice became morbid, and before the mice died of ascites, a decrease in bodyweight was recorded. Based on this data, days 3, 9, and 15 were chosen todetermine changes in immunocompetency in the treated and untreated TBM ascompared with control normal mice.

In Vitro Proliferation and Cytokine Production Assays. On days 3, 9,and 15 after injecting mice with Meth A cells, spleens were removed from theTBM treated with IL-18 or untreated, and control non-tumor-bearing mice(n = 3 mice/group). Single cell suspensions were obtained in loose-fitting

ground glass homogenizers. Erythrocytes were removed by hypoosmoticshock, and the remaining spleen cells were washed twice with cold 0% FBSRPMI 1640 and once with complete medium. Cells were resuspended incomplete medium containing 2.5 jig/ml Con A and adjusted to a density of1 x IO7 cells/ml. To determine proliferation in the presence of a T-cell

mitogen, 100-/J.1 aliquots of the cell suspensions were seeded per well intriplicate in flat-bottomed 96-well plates, and 100-/¿1aliquots of medium only

or medium containing various concentrations of Con A were added. Plateswere incubated at 37°Cfor 72 h and pulsed with 0.5 ¿tCiof [3H]thymidine/well

for the last 18 h of culture. Cells were harvested, and incorporated thymidinewas determined on an automated direct beta counter (Becton Dickinson.Bedford. MA). To determine cytokine production, 2-ml volumes of the samecell suspensions were seeded in triplicate in 12-well culture plates, and the

cells were cultured for 24 h. Culture supernatants were harvested, passedthrough 0.22-fj.m filters, and stored at -20°C until assay for cytokine levels by

ELISA.Flow Cytometric Analysis. Erythrocyte-free spleen cell suspensions of the

treated and untreated normal and tumor-bearing groups were obtained as

described elsewhere in this study. Cells were washed twice with cold PBS andstained with antibodies against mouse CD3 (Cedarlane. Ontario, Canada), CD4(Cedarlane), CDS (Cedarlane), and asialo GM1 (Wako Fine Chemicals, Osaka,Japan) for 45 min on ice. Cells were washed again with cold PBS andincubated with FITC-conjugated secondary antibodies in the dark. After washing again with PBS. cells were examined for positive staining by flow cytom-

34

32

30

3 28

'S) 26

I 24X ' '

1 22

20

18

control mice (n=5)treated mice (n=5)

éter(Epics Profile, Hialeah, FL), and the percentage of positively stained cellswas determined by the cytometer.

MLTC and Cytotoxicity Assays. To determine specific CTL activity,Meth A cells were washed and suspended in medium at a density of 1 X IO6

cells/ml in complete medium. MMC was added to the culture at a concentration of 100 Mg/ml, and the cells were incubated for at least an hour, after whichcells were washed three times with medium and resuspended at 4 X IO5

cells/ml. One-nil volumes of the treated Meth A cell suspensions were seededin 12-well plates. Next, erythrocyte-free spleen cells were obtained as described above and added to the wells at a cell density of 2 X IO7 cells/ml in

1-ml volumes of complete medium containing 20 units/ml IL-2 (final concen

tration, 10 units/ml). Plates were incubated for 5 days, and spleen cells werecollected, washed, and resuspended as 1-ml volumes of 1 X IO7cells/ml. Cells

were also treated with either anti-CD4 or anti-CD8 monoclonal antibodies for30 min on ice and then treated with a final 10-fold dilution of low-toxicityrabbit complement for 30 min at 37°Cbefore washing. To determine cytotoxic

activity, the JAM test was used (12). Briefly. Meth A cells were labeled at2 X IO6 cells/ml with 5.0 ¿tCi/ml[3H|thymidine for 4 h and washed free of the

isotope. Effector and target cells were seeded in triplicate in a final volume of200 ¿ilin round-bottomed well plates at various E:T ratios, and plates were

incubated for 18 h before cell harvest and measurement of thymidine incorporation. The percentage of DNA fragmentation was calculated from triplicatewells according to the formula below:

[(Maximum cpm —¿�sample cpm) 1X 100%

Maximum cpm |

where maximum cpm is the amount of label retained by Meth A cells culturedalone, and sample cpm is the amount of label that was retained by Meth A cellscocultured with effector cells.

Kinetics of NK Cell Activity. Erythrocyte-free spleen cell suspensions

were obtained from normal mice and treated or untreated TBM on the indicated days and seeded in 96-well round-bottomed culture plates containing51Cr-labeled YAC-1 cells at various E:T cell ratios. Plates were incubated for

4 h, after which three control wells containing labeled YAC-1 cells received 2N

HC1 to lyse the cells (maximum release), and three control wells also containing labeled YAC-1 cells were left untreated (spontaneous release). Plates werespun on a centrifuge to sediment the cells, and 100 /xl of cell-free supernatants

were collected, and the amount of label released in the supernatants wasdetermined by an automated counter. The percentage of cytotoxicity wasdetermined according to the following formula:

Percent cytotoxicity =T (Sample cpm - spontaneous cpm) 1

[(Maximum cpm —¿�spontaneous cpm)JX 100%

12 15 18

Effects of in Vivo Treatment with IL-18 on Normal Mice. To determinewhether the effects observed in TBM pretreated with IL-18 can also beobserved in normal mice, we treated mice with 1 /xg of IL-18 i.p. once and

again 3 days later, and we harvested spleens from the treated mice and normalmice 6 days after the first injection of IL-18. Erythrocyte-free spleen cell

suspensions were prepared, and proliferative responses, cytokine production,and NK activity were examined as described in the preceding sections of this

paper.Statistical Evaluations. Degrees of significance between the groups tested

were determined by Student's t test. All experiments were performed at least

twice and were reproducible.

Meth Atransplantation "

days after transplantation

spleen cell harvest

day 3 day 9 day 15

Fig. 1. Treatment protocol for investigation of changes in immunocompetency inTBM. Mice were treated i.p. with IL-18 twice, 3 days and 6 h before i.p. injection withMeth A on day 0. Spleens were harvested on the indicated days, and spleen cells weresubjected to various assays.

RESULTS

Changes in the Proliferative Response to Con A of Untreatedand IL-18-treated TBM. Proliferation of mitogen-stimulated TBM

spleen cells did not differ greatly from that of normal mouse spleencells on days 3, 9, and 15 after tumor transplantation; however, in theabsence of mitogen, day 3 pretreated TBM spleen cells proliferatedabout 13-fold of the levels observed in untreated TBM and normal

mice. On day 9 after tumor transplantation, pretreated TBM spleencell proliferation was still significantly enhanced in the absence ofmitogen when compared to that of normal controls (Fig. 2). Untreated

4558

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTITUMOR KILLER CELL ACTIVATION BY IL-18

days

day 9

day 15

100000

10000

1000

100

100000

loooo

1000

100

100000

10000

1000

1002.5 1 0

Con A ( ng/ml)

•¿�normal mice 0 untreated IBM D IL-18-treated IBM

Fig. 2. Proliferation of normal mouse and treated or untreated TBM spleen ceils in thepresence or absence of various concentrations of the miiogen Con A. *. P < 0.01 versusthe normal mouse and untreated TBM groups; **, P < 0.01 versux the normal mousegroup in the absence of Con A.

TBM spleen cells also exhibited enhanced proliferation, but this waslower than that observed in pretreated TBM. On day 15, proliferationof both treated and untreated TBM spleen cells was significantlygreater than that observed in controls; however, infiltration of untreated TBM spleens by Meth A cells could be observed, and theenhanced proliferation in untreated TBM cultures on day 15 may bepartly due to infiltrating tumor cells.

Changes in in Vitro Cytokine Production by Mitogen-stimu-lated Spleen Cells. Con A-stimulated spleen cells from day 3 TBMpretreated with IL-18 showed suppressed production of IL-2 andIFN-y and a simultaneous significant increase in the production ofIL-10 during 24 h of culture as compared with normal mouse spleencell cultures (Fig. 3). The production of IL-2 almost recovered by day

9 and was fully recovered on day 15 in the pretreated group. Theproduction of IL-10 by pretreated TBM spleen cells decreased on days

9 and 15 in inverse correlation with an increase in IL-2 production but

remained above control levels. There was also a moderate but significant increase in the production of IL-10 in untreated TBM ConA-stimulated spleen cell cultures as compared to that of controls ondays 9 and 15 and a 25% decrease in the production of IL-2 that lastedthroughout the observation period. The production of IFN-y was

decreased in treated and untreated TBM cultures after tumor transplantation, whereas the production of GM-CSF in the pretreated group

was significantly increased above control levels (Fig. 3). The averageconcentrations of cytokines produced by mitogen-stimulated normalmouse spleen cells during 24 h in culture are as follows: IL-2, 16.5ng/ml; IL-10, 600 pg/ml; IFN--y, 72 lU/ml; and GM-CSF, 133 pg/ml.

Changes in the Percentages of Spleen Cell Subsets Induced by11-18. After staining for different mouse spleen cell subsets, a de

crease in the percentages of normal or day 3 TBM spleen cellsexpressing CD3 was observed by flow cytometry after pretreatmentwith IL-18 in vivo when compared with control untreated cells (Table1), which was also reflected in the percentages of CD4- and CD8-

expressing cells. A concurrent increase in the percentage of asialoGM1 + cells was also observed in the treated mouse spleens, indicat

ing that the proliferating cells causing splenomegaly on day 3 after invivo treatment with IL-18 were probably NK cells. On days 9 and 15after transplantation, the percentages of asialo GM1 * cells in the

treated TBM group returned to approximately normal levels, as didthose of other spleen cell subsets (data not shown).

Kinetics of CTL Activity. On day 3 after transplantation of thetumor cells, the spleen cell cytotoxic activities of untreated TBM andTBM treated with IL-18 were approximately equal to that observed in

normal mouse cultures (Fig. 4). The cytotoxic activity observed inuntreated and pretreated TBM could be augmented by treating thecells after MLTC with anti-CD8 monoclonal antibody and complement, indicating that CD8 ' T cells may be exerting a suppressive

effect on cytotoxic antitumor spleen cells in TBM. The cytotoxicactivity in pretreated TBM spleens was significantly enhanced on day9 after tumor injection and was about double that observed in normalmice. Even untreated TBM exhibited a cytotoxic activity higher thanthat observed in normal mouse spleen cell cultures on day 9. Resultsobtained for day 15 were similar to those obtained on day 9 aftertumor challenge. Subset depletion experiments showed that on days 9and 15, the cytotoxic activity in pretreated TBM spleens was due toboth CD4 + and CD8+ cells. The data shown in Fig. 4 represent results

obtained at E:T ratios of 100:1 for all experiments.Kinetics of NK Activity. Our previous study has shown that when

TBM pretreated with IL-18 were depleted of asialo GM1 cells with

antibody early after tumor transplantation, the antitumor effects ofpretreatment with IL-18 were abolished. NK activity of pretreated

TBM on day 3 was significantly enhanced as compared to that ofnormal mice and untreated TBM (Fig. 5). NK activity of the pretreated mice decreased to normal levels on days 9 and 15 aftertransplantation, indicating that the augmentation of NK activity invivo is short-lived. In our experiments, spontaneous release of 5lCr

was always less than 10% of the maximum release.Changes in Proliferation, Cytokine Production and NK Activity

of Normal Mouse Spleen Cells after Treatment with IL-18 in Vivo.The proliferation of spleen cells from normal mice treated with IL-18

in vivo was significantly enhanced in the absence of mitogen whencompared with that observed in control mouse spleens (Fig. 6). Thisenhanced proliferation seemed to be a maximum, because addition ofincreasing amounts of the mitogen Con A could not increase proliferation of the cells. Mitogen-stimulated normal mouse spleen cellsalso exhibited significantly increased productions of IL-10 and GM-CSF and decreased production of IL-2 (Fig. 7). There was no difference in the production of IFN-y between the treated and control

4559

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTITUMOR KILLER CELL ACTIVATION BY IL-18

150 T10000

*

^.2au

1000

" 100cV

So.

dayS day 9 day 15

IL-10

150 r

day 3 day 9 day 15

GM-CSF

days day 9 day 15 " day 3 ¿ay

•¿�normal mouse 0 untreated TBM D IL-18-treated TBM

day 15

Fig. 3. Cytokine production by mitogen-stimulated normal and treated or untreated TBM spleen cells. Spleen cells were stimulated as described in the text, and cytokine levels inculture supernatants were determined by sandwich ELISA. Note that the X axis for IL-10 production is in the logarithmic scale. P < 0.05 for the IL-18-treated group versus the normalmouse and untreated TBM under the same conditions for all the cytokines tested, except for IFN-y and GM-CSF values between the IL-18-treated and untreated TBM.

groups. As shown in Fig. 8, spleen cells of normal mice treated withIL-18 in vivo also exhibited augmented NK activity against YAC-1

cells in vitro. These results were basically similar to those observed inTBM pretreated with IL-18 but are less pronounced in normal miceand indicate that the immunoaugmenting effects of IL-18 in this groupare intrinsic to IL-18 but are possibly amplified by the transplantation

of Meth A cells.

DISCUSSION

We have previously shown that pretreating mice with IL-18 pro

tects them from the lethal effects of i.p. injection with syngeneic MethA sarcoma (11). In the same report, we showed that NK cells play anearly and indispensable role in the protective antitumor effects induced by IL-18, and this NK activity was later followed by theinduction of CD4+ T-cell-mediated memory specific to Meth A. In

this study, we have further examined the effects of IL-18 in Meth A

TBM during the early stages of tumor rejection and compared these

Table 1 Changes in mouse spleen cell subsets after treatment with IL-18

Spleen cell subsets were determined as described in the text using a flow cytometer anda standard gate program. Results represent data from two or more experiments, andnonspecific staining in controls incubated with the FITC-conjugated second antibodies

alone was always less than 3%.

MouseNormalNormalTBMTBMIL-18NoYesNoYesDaysafter tumor

transplantation33Expressing

cells(%)CD329203122CD420102014CDS117107Asialo

GM125383038

with the same immunological parameters in untreated TBM, whichsuccumb to progressive tumor growth within about 3 weeks aftertransplantation, and normal untreated mice.

NK activity was significantly enhanced in the pretreated groupearly after the administration of IL-18 and transplantation of the tumor

but decreased to normal levels by day 9 after tumor injection. On day3, no specific CTL activity could be observed in pretreated TBM, butdepletion of CD8+ cells from the spleen cell cultures enhanced spleen

cell cytolytic activity in the tumor-bearing groups. Production of IL-2by Con A-stimulated pretreated TBM spleen cells was significantlyinhibited on day 3, and because several reports have shown that CD8+

suppressor T-cell-derived IL-10 can suppress IL-2 production (13,14), we next examined the production of IL-10 in mitogen-stimulatedpretreated TBM spleen cell cultures. Indeed, the production of IL-10in these cultures was found to be enhanced 10-fold as compared to

normal control levels, whereas the levels of the cytokine in untreatedTBM cultures also significantly increased. IL-10 production was also

found to be significantly increased in MLTC culture supernatants oftreated and untreated TBM spleen cells compared to those containingnormal mouse spleen cells (data not shown). This indicates thatstimulation of TBM spleen cells with Meth A cells is enough toinduce IL-10 production that is enhanced by treatment in vivo withIL-18. However, the cellular origin of the IL-10 observed in ourmodel is still unknown. The production of IFN--y by mitogen-stimu

lated mouse spleen cells, which is greatly enhanced after exposure toIL-18 in vitro (6), was inhibited in treated and untreated TBM spleencell cultures after tumor transplantation. GM-CSF production was

enhanced in the pretreated TBM cultures to a peak level on day 3and started to decline on days 9-15 after transplantation. Mitogen-

4560

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTITUMOR KILLER CELL ACTIVATION BY IL-18

normal mouse

day 15

untreated TBM IL-18-treated TBM

45

40

£ 35S_^

§ 30VS 25

¡ 20Ulâ„¢¿�15

14—

< 10

Q

normal mouse untreated TBM IL-18-treated TBM

•¿�none 0 anti-CD4 ffl anti-CD8

Fig. 4. Kinetics of CTL activity in normal and treated or untreated TBM spleens.Spleen cells were stimulated with MMC-treated Meth A cells as described in the text for5 days. Cells were treated with anti-CD4 or anti-CD8 antibodies and complement, andcytotoxic activity against fresh isotope-labeled Meth A cells was determined. The amountof thymidine label incorporated into the cells was in the region of 4000 cpm. *, P < 0.01and P < 0.05 versus the normal mouse and untreated TBM group, respectively:**, P < 0.05 versus the IL-18-treated group without antibody treatment.

stimulated spleen cells from untreated TBM produced slightlyincreased amounts of GM-CSF on day 9 after transplantation, which

decreased to below control levels on day 15.Spleen cells from the pretreated mice exhibited enhanced prolifer

ation in vitro in the absence of mitogen when compared to normalmouse spleen cells. This enhanced proliferation did not diminish up to2 weeks after treatment with IL-18 in vivo. The normal mouse and

TBM spleen cells causing splenomegaly in the early stages aftertreatment with IL-18 were identified as asialo GM1+ cells by flow

cytometry. Spleen cell subsets were found to return to normal levelson days 9 and 15 after treatment with IL-18 (data not shown). A recentreport has shown that IL-12 preferentially stimulates a subset of asialoGMI+ CD8+ mouse spleen T cells (15). In our system, IL-18 in

creased the percentage of asialo GM1 *"cells but slightly decreased the

percentage of CDS-expressing cells on day 3 after challenge. Therefore, we consider that IL-18 probably induces the proliferation ofasialo GM1+ CDS" cells in the early stages after tumor challenge.

The untreated TBM spleens also contained proliferating cells on days9 and 15; however, on day 15, TBM spleens were found to beinfiltrated with Meth A cells, therefore the results on spleen cellproliferation in this group cannot be interpreted with certainty.

The enhanced NK activity, cytokine production profile, and enhanced unstimulated spleen cell proliferation observed in the pretreated TBM group were also observed in normal mice treated twicewith IL-18 and sacrificed 3 days after the last administration of IL-18.Untreated mitogen-stimulated as well as MMC-treated Meth A-stim-ulated untreated TBM spleen cells (5-day MLTC cultures) also produce increased amounts of IL-10 when compared to control normal

mouse spleen cell cultures (Fig. 3 and data not shown), indicating that

15

ioo

tuo.

day 15

200 12.5

•¿�normal mouse 0 untreated TBM D IL-18-treated TBM

Fig. 5. NK activity in fresh spleen cells harvested from normal mice and treated oruntreated TBM. NK activity in spleen cells harvested on day 3 was significantly enhancedin the TBM group treated with IL-18 when compared to normal mouse and untreatedTBM activity (P < 0.05) but relumed to control normal mouse levels on days 9 and 15.

4561

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTnUMOR KILLER CELL ACTIVATION BY IL-18

the immune system is activated in the Meth A tumor-bearing state andenhanced by treatment with IL-18. Because IL-10 is also known toinduce NK activity in mice ( 16), it is possible that IL-18, which is alsoknown to directly augment NK activity, cooperates with IL-10 to

produce a stronger antitumor effect. Cooperation between these twocytokines can take the form of an additive effect on NK cell activationor a direct effect on the tumor cells, because IL-10 has been reportedto confer sensitivity on NK-resistant tumor cells ( 17, 18). We have notattempted to use neutralizing antibodies against IL-10 to determine

whether this cytokine is involved in the antitumor effects induced byIL-18, mainly because of the large amounts of antibody reportedly

required to neutralize the cytokine in vivo (19). It is interesting,however, that a decrease in the production of IL-10 by treated TBM

spleen cells is accompanied with a simultaneous drop in NK activityto normal levels.

In a previous in vitro study on the stimulatory effects of humanIL-18, we demonstrated that IL-18 inhibits the production of IL-10 inCon A-stimulated cultures of human peripheral blood mononuclear

cells, even though NK activity was stimulated in similar cultures inthe absence of mitogen (7). We also showed in a related study thatIL-18 enhances IL-2 production by enriched human T cells stimulatedwith anti-CD3 antibody (9). The apparent discrepancies between our

previous in vitro results and the results revealed in this study are

100000

& 10000

1000

100

•¿�untreated miceD IL-18-treatedmice

1.25 0.625

Con A iig/ml

Fig. 6. Proliferation of spleen cells in the absence of milogen is also observed in normalmice injected twice with IL-18. Mice were treated with the cylokine according to the sameprotocol as the IL-18-treated TBM mice, except that they were not transplanted withtumor cells. *. P < 0.005 versus the untreated mice.

40

30

I z°

Iu

10

•¿�untreated miceD IL-18-treatedmice

200 100 50

E:T ratio

25 12.5

Fig. 8. NK activity of normal mice after treatment with IL-18. Mice were treated twicewith IL-18. and spleen cells were harvested 3 days after the last administration of IL-18.NK activity was determined against YAC-1 in a 4-h MCr release assay. *, P < 0.05 versus

the untreated control mouse spleen cells.

probably due to the different environments encountered in the in vivoand in vitro systems. These discrepancies are not a result of thedifferent species examined in these studies, because Kohno et al. (8)have also recently shown that murine IL-18 enhances IL-2 production

by Th type 1 clones in vitro. Therefore, we think that the complexinteracting immune networks active in vivo are the cause for thediscrepancies between results obtained in vitro and in vivo.

In conclusion, IL-18 stimulates NK activity in vivo, possibly incooperation with other cytokines such as IL-10, followed by the

generation of CTL. The sequential generation of a nonrestrictedeffector (NK) cell response followed by a specific (CTL) responseinduced by IL-18 in vivo provides a useful model for experimental

treatment protocols with emphasis on the timing of treatment. Thisstudy will help identify novel molecules involved in in vivo networksof antitumor responses and immunological parameters that indicatethe possibility of a beneficial outcome after treatment with immuno-stimulatory cytokines such as IL-18.

ACKNOWLEDGMENTS

We are indebted to Drs. K. Ohashi. K. Iwaki, S. Arai, and M. Takeuchi ofour institute for scientific discussion and assistance throughout this study. Weare also indebted to K. Torigoe and M. Fujii for expression of IL-18 cDNA andproduction of murine IL-18.

.1

400

300

200

100

•¿�untreated miceD IL-18-treated mice

IL-2 IL-10 GM-CSF

cytokine

Fig. 7. Cylokine production by normal mouse spleen cells injected twice with IL-18 asdescribed in the text. Cytokine production in normal and TBM injected with 1L-18according to the same protocol followed a similar pattern. *, P < 0.005 versus theuntreated mouse culture supernatants.

4562

REFERENCES

1. Fu, Y-X.. Watson. G. A-, Kasahara. M.. and Lope/. D. M. The role of tumor-derived

cytokines on the immune system of mice hearing a mammary adenocarcinoma. I.Induction of regulatory macrophages in normal mice by the m vivo administration ofrGM-CSF. J. Immunol., 146: 783-789. 1991.

2. Loeffler, C. M., Smyth. M. J.. Longo. D. L., Kopp, W. C.. Harvey. L. K., Tribble,H. R., Tase. J. E., Urba, W. J.. Leonard. A. S.. Young. H. A., and Ochoa. A. C.Immunoregulation in cancer-bearing hosts: down-regulation of gene expression andcytotoxic function in CD8* T cells. J. Immunol.. 149: 949-956. 1992.

3. Xu, Z-Y., Hosokawa. M.. Morikawa. K.. Hatakeyama. M.. and Kobayashi. H.Overcoming suppression of untitumor immune reactivity in tumor-hearing rats bytreatment with bleomycin. Cancer Res.. 48: 6658-6663. 1988.

4. Young. M. R. I.. Halpin. J., Wang, J.. Wright, M. A.. Matthews. J., and Schmidt Pak.A. la,25-Dihydroxyvitamin Dt plus 7-interferon blocks lung tumor production ofgranulocyte macrophage colony-stimulating factor and induction of immunosuppres-sor cells. Cancer Res.. 53: 6006-6010. 1993.

5. Micallef. M. Immunoregulatory cytokine production by tumor-bearing rat spleen cellsand its modulation by bleomycin. Anticancer Drugs, 4: 213-222. 1993.

6. Okamura, H.. Tsutsui, H.. Komatsu. T.. Yutsudo. M.. Hakura, A.. Tanimoto, T..Torigoe. K.. Okura. T.. Nukada. Y.. Hattori. K., Akila. K.. Namba, M.. Tanabe. F.,Konishi. K.. Fukuda. S.. and Kurimoto. M. Cloning of a new cytokine that inducesIFN-y production by T cells. Nature (Lond.), 378: 88-91, 1995.

7. Ushio. S., Namba, M.. Okura, T.. Hattori. K.. Nukada. Y.. Akita, K., Tanabe, F.,

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

ANTITUMOR KILLER CELL ACTIVATION BY IL-18

Konishi. K.. Micallef. M.. Fujii, M.. Torigoe, K.. Tanimolo. T.. Fukuda. S.. Ikeda. M..Okamura. H.. and Kurimolo. M. Cloning of the cDNA for human IFN-y-inducingfactor, expression in Est'tierichia culi, and studies on the biologic activities of the

protein. J. Immunol., 156: 4274-4279, 1996.

8. Kohno, K.. Kataoka. J.. Ohlsuki. T.. Suemoto. Y.. Okamolo, I.. Usui. M.. Ikeda. M.,and Kurimoto. M. IFN-y-inducing factor (IGIF) is a costimulatory factor on theactivation of Th I but not Th 2 cells and exerts its effect independently of IL-12.J. Immunol.. 158: 1541-1550. 1997.

9. Micallef, M. J.. Ohtsuki. T.. Kohno. K.. Tanabe. F.. Ushio. S., Namba. M.. Tanimolo.T., Torigoe. K., Fujii. M., Ikeda. M., Fukuda. S.. and Kurimoto. M. Inlerferon-y-inducing factor enhances T helper I cytokine production by stimulated human T cells:synergism with intcrleukin-12 for interferon-y production. Eur. J. Immunol.. 26:1647-1651. 1996.

10. Tsulsui. H.. Nakanishi. K.. Matsui. K.. Higashino. K.. Okamura. H.. Miyazawa. Y..and Kaneda. K. IFN-y-inducing factor up-regulales Fas ligand-mediated cytotoxicactivity of murine natural killer cell clones. J. Immunol.. 157: 3967-3973. 1996.

11. Micallef, M. J., Yoshida. K., Kawai. S.. Hanaya. T.. Kohno, K., Arai, Y.. Tanimoto.T.. Torigoe. K.. Fujii, M.. Ikeda. M.. and Kurimoto, M. In vivo antitumor effects ofmurine intcrferon-y-inducing factor/interleukin-18 in mice bearing syngeneic Meth Asarcoma malignant ascites. Cancer Immunol. Immunother.. 43: 361-367. 1997.

12. Malzinger. P. The JAM test—a simple assay for DNA fragmentation and cell death.J. Immunol. Methods. 145: 185-192. I99I.

13. Inoue. T.. Asano. Y.. Matsuoka. S.. Furutani-Seiki. M.. Aizawa. S.. Nishimura. H..Shirai. T.. and Tada. T. Distinction of mouse CDS * suppressor effector T-cell clones

from cytotoxic T-cell clones by cytokine production and CD45 isoforms. J. Immunol..ISO: 2121-2128. 1993.

14. Nanda. N. K.. Sercarz, E. E.. Hsu, D-H.. and Kronenberg. M. A unique pattern oflymphokine synthesis is a characteristic of certain antigen-specific suppressor T-cellclones. Int. Immunol.. 6. 731-7373, 1994.

15. Lee, U., Santa. K.. Habu. S.. and Nishimura. T. Murine asialo GMI * CDS* T cells

as novel interleukin-12-responsive killer T cell precursors. Jpn. J. Cancer Res.. 87:429-432. 1996.

16. Zheng. L. M.. Ojcius. D. M.. Garaud. F.. Roth. C, Maxwell, E.. Li. Z.. Rong. H.,Chen. J.. Wang. X. Y., Catino, J. J.. and King, I. Interleukin-10 inhibits tumormetastasis through an NK cell-dependent mechanism. J. Exp. Med.. 184: 579-584.

1996.17. Carson. G. E.. Lindemann. M. J., Baiocchi. R.. Linen. M.. Tan. J. C.. Chou. C-C.,

Narula. S.. and Caligiuri. M. A. The functional characterization of interleukin-10receptor expression on human natural killer cells. Blood. «5:3577-3585. 1995.

18. Salazar-Onfray. F.. Petersson. M.. Franksson. L., Matsuda. M.. Blankenstein. T..Karre. K.. and Kiessling. R. IL-10 converts mouse lymphoma cells to a CTL-resistant.NK-sensitive phenotype with low hut peptide-inducible MHC class I expression.J. Immunol.. 154: 6291-6298. 1995.

19. Giovanelli. M.. Musiani. P.. Modesti. A.. Dellabona. P.. Castorati. G., Musso. T., andForni. G. Local release of IL-10 by transfected mouse mammary adenocarcinoma cell

does not suppress but enhances antitumor reaction and elicits a strong cytotoxiclymphocyte and antibody-dependent immune memory. J. Immunol.. 155: 3112-3123.

1995.

4563

Association for Cancer Research. by guest on August 27, 2020. Copyright 1997 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from