(CANCER RESEARCH 58, 2838-2843. July I. 1998]

Effective Adoptive Immunotherapy by T-LAK Cells Retargeted with Bacterial

Superantigen-conjugated Antibody to MUC1 in Xenografted SevereCombined Immunodeficient Mice1

Masao Shinoda, Toshio Kudo,2 Masanori Suzuki, Yu Katayose, Naoki Sakurai, Hisaaki Saeki, Hideaki Kodama,

Kenji Fukuhara, Kohzoh Ima!, Yuji Hinoda, and Seiki MatsunoFirst Department of Surgen', TohtìkiiUniversity School of Medicine ¡M.Sh., M. Su., Y. K., N. S., H. K., K. F., S. M.] and Cell Resource Center for BiomédicalResearch. Instituteof Development. Aging, and Cancer. Tohoku University, {T. K.. H. S. l, Sendai 980-8575. and Department of Intentai Medicine. Sapporo Medical Universit\ School of Medicine,S-l, W-16. Chuo-ku. Sapporo 060-0061 [K. L, Y. H.¡,Japan

ABSTRACT

To reinforce cytotoxic activity and the targeting ability of lymphokine-activated killer cells with a T-cell phenotype (T-LAK) for adoptive im-

munotherapy against human bile duct carcinoma (BDC), staphylococcalenterotoxin A (SEA) was conjugated chemically with MUSEI 1 monoclonal antibody (MUSEI 1 mAl»,directed to the MUC1 antigen, usingV-smriiimiiiKI 3-(2-pyridyldithio) propionate and 2-iminothiolane HCI.Both SEA-conjugated MUSEI 1 niAli (SEA-MUSE11) and the F(ab')2 ofMUSE11 niAh (SEA-F(ab')2) showed significant enhancement of T-LAK

cell tumor neutralization for MUC1 positive-target tumor cells, even with

a concentration of 0.01 fig/ml at an E:T ratio of 5:1 in vitro. In this m vitrotest, MUCl-positive BDC cells were observed to attach to surroundingT-LAK cells in the presence of SEA-MUSE11 or SEA-F(ab')2. Remarka

ble tumor growth inhibition was observed in BDC-grafted severe combined immunodeficient mice to which 2 x 10 T-LAK cells preincubatedwith 2 ftg of SEA-MUSE11 or SEA-F(ab')2, together with recombinant

interleukin 2 (500 IU), were administered i.v. for 4 consecutive days, whentumor size was 5 mm in diameter. These results point to a promisingadoptive immunotherapy for patients with BDC.

INTRODUCTION

The prognosis of patients with BDC3 is known to be poor because

of the difficulty of curative resection due to irregular tumor invasionto lymphatic tissues and/or perineural spaces (1). Even adjuvantchemotherapy or radiotherapy after surgery generally cannot achievesatisfactory results. To improve this situation, immunotherapy hasattracted a great deal of attention. In a previous study of humanBDC-grafted SCID mice, specific targeting therapy, consisting of i.v.administration of T-LAK cells sensitized with two kinds of bispecificantibodies, i.e., anti-CD3 X anti-MUCl and anti-CD28 X anti-MUCl

antigen (2), demonstrated remarkable inhibition of tumor growth.However, complete cures could not be obtained. To develop anotherstrategy that could be used together with this specific targeting therapy, we have concentrated our attention on SAgs. SAgs, named byWhite et al. (3), are the most potent known activators of human Tlymphocytes. They bind outside of the peptide-binding groove of

MHC class II molecules and activate T cells expressing a certain Vß

Received 10/13/97; accepted 5/1/98.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 Grant-in-Aid for Cancer Research 06279101

from the Ministry of Education. Science, Sports and Culture of Japan.- To whom requests for reprints should be addressed, at the Cell Resource Center for

BiomédicalResearch. Institute of Development, Aging and Cancer. Tohoku University,4-1 Seiryomachi. Aoba-ku. Sendai 980-8575. Japan. Fax: 81-22-717-8573.

' The abbreviations used are: BDC. bile duct carcinoma; SAg. superantigen; TCR,

T-ccll receptor: mAb. monoclonal antibody; Ah. antibody; LAK. lymphokine activatedkiller: T-LAK, LAK with T-cell phenotype; SEA, staphylococcus enterotoxin A; IL-2,recombinant human interleukin 2: SCID, severe combined immunodeficienl: SPDP.W-succinimidyl 3-(2-pyridyldithio) propionate: SEA-MUSE11. SEA-conjugatedMUSEI 1 mAb; SEA-F(ab'),, SEA-conjugated F(ab'), of MUSEI 1 mAb; TN, tumor

neutrali/ation; MTS, (3-(4.5-dimethylthiazole-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2//-tetrazolium, inner salt.

type of TCR, unlike the case with conventional antigen recognition(4). To use this potential for adoptive immunotherapy, SEA andMUSEI 1 mAb directed to the epithelial mucin antigen MUC1, widelyexpressed in adenocarcinomas, were conjugated chemically in thisstudy. SEA binds T-LAK cells through MHC class II molecules and

thereby activates them. MUSEI 1 mAb, on the other hand, increasesthe ability of T-LAK cells to target MUC1 after introduction of anappropriate ligand. We therefore speculated that SEA-conjugatedMUSEI 1 mAb (SEA-MUSE 11) might enhance the specific cytotoxicactivity of T-LAK cells against MUCl-positive tumors. To test this

hypothesis, in vitro and in vivo experimental studies were carried out.In this study, we report the first effective adoptive immunotherapy byhuman T-LAK cells with SEA-conjugated antibodies in BDC solidtumor-grafted SCID mice.

MATERIALS AND METHODS

inAlis. For induction of T-LAK cells and construction of SEA-MUSE 11,the following mAbs were used: OKT-3 (anti-CD3, mouse IgG2a) andMUSEI 1 (anti-MUCl. mouse IgGl) produced by Hinoda (5). These mAbs

were purified from sera and ascitic fluid of mice inoculated with hybridomacells by caprylic acid treatment in combination with the ammonium sulfatemethod (6). For the flow cytometry analysis, W6/32 (anti-MHC class 1), L243(anti-MHC class II), WT31 (anti-TCR-a/3; Becton Dickinson. San Jose, CA),11F2 (anti-TCR--yS; Becton Dickinson), and MUSEI I mAbs were used as the

first antibodies, and FITC conjugated goat anti-mouse IgG (6250, Biosource

International, Camarillo, CA) was used as the second antibody.Cell Lines. The human BDC cell line (TFK-1). reactive with MUSE 11

mAb, was used as a target and a human hepatocellular carcinoma cell line(HT-17) as a control. Both were established in our laboratory (7) and culturedin RPMI-1640 supplemented with 10% fetal bovine serum, 100 units/ml

penicillin, and 100 /¿g/mlstreptomycin (designated as culture medium below).Reagents. SEA (AP101) was purchased from Toxin Technology, Inc.

(Sarasota. FL). Rabbit anti-SEA Ab (S-7656) was purchased from SigmaChemical Company (St. Louis. MO). Recombinant human IL-2 was kindly

supplied by Shionogi Pharmaceutical Co. (Osaka. Japan).Effector Cells. For induction of T-LAK cells, peripheral blood mononu-

clear cells, isolated by density gradient centrifugation from a healthy volunteer,were cultured for 48 h in culture medium supplemented with 100 lU/mlrecombinant human IL-2 at a cell density of 1 X lO^/ml in a culture flask (A/S

NUNC, RosKilde. Denmark) precoated with OKT-3 mAb (10 /xg/ml). TheT-LAK cells were then transferred to another flask and expanded in culturemedium containing 100 lU/ml IL-2 for 2-3 weeks.

Production and Purification of SEA-conjugated MUSE11 and Hah'K

of the MUSE11 Ab. The MUSEI 1 mAb was purified from sera and asciticfluid of mice inoculated with hybridoma cells and its F(ab'), was obtained by

digestion with preactivated papuin in 0.1 M acetate buffer (pH 5.5) at 37°Cfor

6 h at a concentration of 15 mg/ml in 40 mM sodium phosphate/150 mMsodium chloride, pH 7.5. For conjugation. mAb or F(ab'), of mAb (12 mg/ml)

was mixed with a 3-fold molar excess of SPDP (Pierce Chemical Co.), which

had been dissolved in DMSO at a concentration of 64 mM and diluted 1:10with 40 mM sodium phosphate/150 mM sodium chloride, pH 7.5 (8). Themixture was incubated at room temperature for 2 h with gentle rocking understerile conditions, then applied to a Sephadex G25-prepacked PD-10 column

2838

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

ADOPTIVE IMMUNOTHERAPY WITH SEA-CONJUGATED ANTIBODY

Table 1 Flow cytomelric analyses of the reactivities ofTFK-l, HT-I7, and T-LAK cells lo antibodies

Reactivities of TFK-I. HT-17, and T-LAK cells to various antibodies and SEA. These results were obtained by flow cytometry using W6/32 (anti-class 1). L243 (anticlass II). WT31(anti-TCRaß). I1F2 (anti-TCR-yS). MUSEI 1 (anti-MUCl), and SEA-F(ab'), of MUSEI I (anti-MUCl) as the first Abs and FITC-conjugated goat antimouse Ig as the second Ab.Binding of SEA, SEA-MUSE11, and SEA-F(ab')2 lo these cells was determined. Cells were first incubated with SEA, SEA-MUSE11, or SEA-F(ab'),. After washing, rabbit anti-SEAas the first Ab and FITC-conjugated goat anlirabbit IgG as the second Ab were applied for staining. With binding tests of SEA-MUSE 11 and SEA-F(ab' )2. similar results were obtained

when FITC-conjugated goat antimouse IgG Ab was used for determination, instead of rabbit anti-SEA as the first Ah and FITC-conjugated goat antirabbit IgG as the second Ab. Figuresin the table are percentages of positively staining cells. Figures in the parentheses are relative mean fluorescence intensities, calculated as the ratios of mean fluorescence intensitiesof test cells to negative controls. Data are from triplicate determinations.

AntigenexaminedMHC

classIMHCclassIITCRa/3TCR-ySMUC1MUC1SEA

bindingSEA-MUSE11bindingSEA-F(ab'),bindingmAbW6/32L243WT311IF2MUSEI

1MUSE11 F(ab'),TFK-1ND"4.03(2.1)NDND92.2

(36.4)88.7(18.8)2.33(1.2)72.2

(42.4)70.1(21.8)Test

cellsHT-17ND1.13NDND1.651.250.821.84NDT-LAK99.9

(99.6)96.0(95.0)93.3(42.4)9.8(9.7)0.840.1099.9(15.3)92.7(10.0)70.5(7.1)

' ND, not done.

(Pharmacia Biotech) equilibrated with 1 mM EDTA-PBS (pH 7.5) for purification of SPDP-IgG or SPDP-F(ab')2 of mAb. SEA at a concentration of 10

mg/ml in 40 mM sodium phosphate/150 mM sodium chloride pH 7.5 was mixedwith a 3-fold molar excess of 2-iminothiolane HCI (Pierce Chemical Co.),

prepared as a 20 mM solution in 50 mM sodium phosphate (pH 8.6), andimmediately incubated for 2 h at room temperature with gentle rocking understerile conditions. Excess reagents and low molecular weight reaction productswere removed by gel filtration on a Sephadex G-25 prepacked PD-10 columnequilibrated with l mM EDTA-PBS (pH 7.5). Both fractionated reactants,SPDP-modifted MUSEI 1 [or F(ab')2] and 2-iminothiolane-derivatized SEA.were mixed at a molar ratio of 1 (SEA):1 [MUSEI I mAb or F(ab')2[. This

mixture was incubated for 2 h at room temperature with gentle rocking and leftat 4°Covernight. The following day, gentle rocking was continued for 4-5 h,

and the reaction mixture was applied to gel filtration chromatography (FPLCsystem) with a HiLoad 16/60 Superdex 200 prep-grade column (Pharmacia

Biotech) equilibrated in 10 mM phosphate/150 mM NaCl (pH 7.5) to removeunconjugated SEA and other materials. In the production of SEA-MUSE 11,

two large peaks of M, 180,000 and M, 210,000, corresponding to 1:1 and 2:1molar ratios of SEA:MUSE11 mAb, respectively, were fractionated. In thecase of conjugating F(ab')2 with SEA, two large peaks of M, 130,000 and M,160,000 were obtained. The purity of SEA-MUSE 11 and SEA-F(ab')2 was

determined by SDS-PAGE. Conjugates with a 1 (SEA):1 [MUSEI 1 mAb orF(ab')2 of MUSEI I mAb] molar ratio were used for the experiments described

below.

Analysis by Flow Cytometry. Flow cytometric analysis was performed ona FACScan flow cytometer using CellQuest software (Becton Dickinson, SanJose, CA).

Binding of SEA, SEA-MUSE11, and SEA-F(ab')2 to TFK-1, HT-17, or

T-LAK Cells. Cells (5 X IO5) were first incubated with 50 /il (5 ¿ig/ml)ofSEA, SEA-MUSE11. or SEA-F(ab')2. After washing, they were exposed to

rabbit anti-SEA as the first Ab and then to FITC conjugated goat anti-rabbit

IgG (H + L, 0833. Immunotech Inc.. Westbrook. ME) as the second Ab. Thestained cells were analyzed by flow cytometry.

TN Assay. TN assay was determined with an MTS assay kit (CellTiter96â„¢AQUEOUS nonradioactive cell proliferation assay: Promega Co. Madi

son, WI; Ref. 9). Target cells, detouched with 0.02% EDTA-PBS solution,were adjusted to I05/ml with culture medium, and 10.000 cells in l(X)-/nl

culture medium were distributed to each well of a half-area (A/2) 96-wellflat-bottomed plate (Costar Corp., Cambridge, MA). They were cultured

overnight to allow adhesion to the wells of the plate, and after removing theculture medium by aspiration. 50 ^1 of various concentrations of SEA-MUSE11, SEA-F(ab')2, SEA. or MUSEI 1 mAb in culture medium and 50 /¿I

of effector cell suspension were added to each well. E:T ratios tested in thisstudy were 2.5 and 5. After culture for 48 h at 37°C.each well was washed

with 100 ¡t\of PBS three times to remove effector cells and dead target cells.This was followed by the addition of 95 /¿I/wellof culture medium and 5

fil/well of a fresh mixture of MTS/phenazine methosulfate solution (PromegaCo.). The plates were read on a microplate reader (Bio-Rad model 3550) at 490

E/T=5

Fig. 1. TN of T-LAK cells with various concentrations of Abs and SEA against TFK-1 cells. Various concentrations of MUSE II mAb, SEA, SEAplus MUSEI 1 mAb (SEAiMUSEll. 1:1). SEA-MUSE11, or SEA-F(ab')2 were added to T-LAK

cells. The TN index was determined by 48-h MTSassay at an E:T ratio of 5:1. Columns show theresults (means; bars, SD) of multiplícate determinations; *. P < 0.001 for 0.01-1 fig/ml versus 0

l for each group (by Student's / test).

C

Co

N

aZ

E3I-

X

0.01 0.05 0.1

Concentration

0.5 1.0

D MUSE11 üSEA SEA+MUSE11(SEA : MUSE11

^SEA-MUSE11

: 1)

SEA-F(ab')2

2839

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

ADOPTIVE IMMl NOÕHKRAPY WITH SEA-CONJUGATED ANTIBODY

Fig. 2. TN of T-LAK cells prestimulated by SEA. T-LAK cells wereincubated at 37°Cfor 6 h with 0.1 ^tg/rnl or 0.5 /ig/ml of SEA. washed,

resuspended in culture medium, and applied to the TN assay (48-h MTSassay; E:T, 5:1 ) of TFK-1 cells. Columns show the results (means; bars,

SD) of multiple determinations.

E/T =5

Ou gìml 0.1 /¿g/ml 0.5 /¿g/ml

nm after incubation for 1-3 h at 37°C.The TN index (2) was calculated asfollows: TN index C7r)= [ 1 - (A4Viof experiment//^, of control)] x 100.

Blocking Tests. To confirmthe specificityof the SEA-MUSE11enhancedTN index, blocking tests using MUSE11 and anti-SEA antibodies were performed. For this, T-LAK cells in the presence of SEA-MUSE11 (0.5 /xg/ml)underwent 48-h MTS assays at an E:T ratio of 2.5, with rabbit anti-SEA Ab(0-30 Mg/ml)or MUSEI l mAb (0-50 /ig/ml) added for blocking.

In Vivo Tumor Models. SCID mice (Fox CHASE C.B.17/Icr-Scid Jcl),6-8 weeks of age, purchased from Japan Clea (Tokyo, Japan), were inoculatedwith 5 x IO6TFK-1 cells s.c. into the dorsal thoracic wall on day 0. The

treatments were initiated on day 10 after tumor inoculation. For 4 consecutivedays, the mice received adoptive immunotherapy, i.e., 2 X IO7T-LAK cells in

a total volume of 0.15 ml were preincubated with 2 fig of either SEA-MUSE11, SEA-F(ab'),. MUSEI 1 mAb, F(ab'), of MUSEI 1 mAb, or SEA at4°Cfor 1 h. Then each xenografted mouse received i.v. via the tail vein 500IU (2 /nl) IL-2 and T-LAK cell suspension (2 x 107/mouse/day)in 0.15 ml of

PBS without washing out antibodies and SEA. Tumor size was measured witha caliper weekly for 10 weeks, and tumor weight (HOin mg was calculatedfrom linear measurements of the width (A) in mm and the length (ß)in mm asfollows (10):

W=(A2XB)/2

Statistical Analysis. Resultswere comparedusing Student'st test.

RESULTS

Reactivities of TFK-1, HT-17, and T-LAK Cells. Data for reactivities of TFK-1, HT-17, and T-LAK cells to various antibodies,

SEA, and SEA-conjugates, as assessed by flow cytometry, are summarized in Table 1. TFK-1 cells had strong reactivities with MUSEI 1,SEA-MUSE11, and SEA-F(ab')2 but were not reactive with SEA.

HT-17 cells, used as a control, had no reactivity with SEA andSEA-MUSE 11 because of their lack of expression of the MUC1

antigen and MHC class II molecules. On the other hand, most T-LAK

cells in this study expressed MHC class I, class II, and TCRaßbutdemonstrated only negligible MUC1 antigens. Consequently, T-LAKcells had strong reactivities with SEA, SEA-MUSE11, and SEA-F(ab')2- The results suggest that both SEA-MUSE11 and SEA-F(ab')2

are able to bind well to TFK-1 and T-LAK cells but not HT-17 cells,indicating that SEA-conjugates are appropriate for retargeting T-LAK

cells.SEA-Antibody-mediated T-LAK Cell TN. Various concentra

tions of MUSEI 1 mAb, SEA, SEA plus MUSEI 1 mAb (mixture ofSEA and MUSEI 1 mAb at 1:1 by molar ratio), SEA-MUSE11, orSEA-F(ab')2 were added to TFK-1 target cells and T-LAK effector

cells in the MTS TN assay (Fig. 1). The TN index of T-LAK cellsalone was —¿�20%,and this was much enhanced (to —¿�60%)by theaddition of SEA-MUSE11 or SEA-F(ab')2. even at 0.01 ¿ig/ml,indi

cating applicability for in vivo experiments in which the dose of SEAshould be kept as low as possible. Addition of the mixture (SEA plusMUSEI 1 mAb) did not enhance TN index statistically, although themean value was slightly elevated.

TN of T-LAK Cells Prestimulated by SEA. To confirm the effectof SEA on T-LAK cells, the TN due to T-LAK cells prestimulated by

E/T=2.5

Fig. 3. Blocking tests with various concentrations of MUSE 11 and anti-SEA Abs. TN of TFK-1cells by T-LAK cells was determined by 48-h MTSassay in the presence of SEA-conjugated MUSEI IAb (0.5 »ig/ml)at an E:T of 2.5:1, with variousconcentrations of rabbit anti-SEA or MUSEI I Absadded. Columns show the results (means; bars. SD)of multiplícate determinations.

oEa

*

SEA-MUSE11anti-SEAMUSE-1

10000.5000.50.10.10.51.01.00.510100.520200.53050

2840

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

ADOPTIVE IMMUNOTHERAPY WITH SEA-CONJUGATED ANTIBODY

<*)

40E/T.5

Fig. 4. TN of HT-17 cells by T-LAK cells. Various concentrations ofSEA-MUSE11 were added to the 48-h MTS assay system (E:T. 5:1 ) withHT-17 target cells. Columns show the results (means; bars. SD) of

multiplícate determinations.

30

£ 20

10 mill0.01 0.05 0.1 o.s 1.0

SEA was measured (Fig. 2). T-LAK cells were preincubated at 37°C

for 6 h with 0.1 or 0.5 fig/ml SEA and then washed and applied forthe TN assay against TFK-1 cells. The direct TN index of T-LAK

cells was slightly increased by this SEA prestimulation, indicating thatthe significant enhancement of T-LAK cell TN in the presence ofSEA-MUSE11 or SEA-F(ab')2 (Fig. 1) was dependent not only on

increased targeting ability but also on direct SEA stimulation ofT-LAK cells, although this increase of T-LAK cell TN with SEA was

nonspecific.Specificity of TN. To verify the specificity of the TN in this study,

blocking tests were performed. When MUSEI I or anti-SEA antibodies were added to the assay system, significant dose-dependent reduction of the TN index was observed (Fig. 3). Enhancement of T-LAKcell TN by SEA-MUSE 11 was not obtained, even at 1.0 /ig/ml, whenthe hepatocellular carcinoma cell line HT-17 was used as a MUC1-

negative target (Fig. 4). These results indicate that the effects ofSEA-MUSE 11 or SEA-F(ab')2 on T-LAK cells are dependent on the

specificity of the MUSEI 1 antibody.Attachment of T-LAK Cells Mediated by SEA-MUSE11.

TFK-1 cells and T-LAK cells were observed during the TN assayusing a phase-contrast microscope (Fig. 5). When SEA-MUSE 11 wasadded to TFK-1 and T-LAK cells in coculture, T-LAK cells started to

Fig. 5. Micrographs of cells in the TN assay. TFK-1 andT-LAK cells were cocultured for 18 h in a A/2 96-wellmicroplate in the presence (A) or absence (B) of SEA-

MUSE1I (l ¿ig/ml;E:T, 5:1).

(ligi ml)

SEA-MUSE11

surround TFK-1 cells after only 1 h. This phenomenon did not occurin the absence of SEA-MUSE 11.

Experimental Treatments of Xenografted SCID Mice. Experimental treatments were initiated 10 days after s.c. inoculation of5 X IO6TFK-1 cells into the dorsal thoracic walls of SCID mice. For

4 consecutive days, xenografted SCID mice received injections i.v. ofIL-2 (500IU) and T-LAK cells (2 X IO7cells) preincubated with 2 /xgof SEA-MUSE11, SEA-F(ab')2, MUSEI 1 mAb, F(ab')2 of MUSEI 1mAb, or SEA at 4°Cfor l h in 0.15 ml PBS. Each group consisted of

15-25 mice. The tumor size was measured weekly for 10 weeks (Fig.6). With both SEA-MUSE11 and SEA-F(ab')2. marked inhibition of

tumor growth was observed from an early phase and continuedthroughout the experimental period. The average inhibition was—¿�70%,but no complete disappearance of tumors was noted.

DISCUSSION

Bacterial SAgs are the most potent known activators of human Tlymphocytes; they bind to the outside of the peptide-binding groove of

MHC class II molecules and present to T cells in an unprocessed form.Only T cells expressing a certain Vßof the TCR recognizing SAgs areactivated, unlike the case with conventional antigen recognition (4).

•¿�-•&-.

2841

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

ADOPTIVE IMMUNOTHERAPY WITH SEA-CONJUGATED ANTIBODY

(g)

2345678

Weeks after treatment

9 10

2345678

Weeks after treatment

9 10

(g) 6

2 5? 4

! 3oE 2D

I- 1

01 23456789 10

Weeks after treatment

Fig. 6. Results of m viro adoptive ¡mmunotherapy of TFK-l tumors in SCID mice.T-LAK cells were incubated with 0.15 ml PBS containing 2 /¿gof SEA-MUSE11 (•).SEA-F(ab'), (A). MUSEI i Ab (Al, F(ab')2 of MUSEI 1 Ab (O), or SEA (D) at 4°Cfor

l h. Then, each mouse received i.v. 500 lU (2 ¡il)IL-2 and 0.15 ml preincubated cellsuspension without washing the cells. Experimental groups receiving T-LAK cells incubated with PBS (*| and PBS control group <*) are included in these figures. Symbolsshow mean values for tumor weights. Bars. SDs of the control. SEA-MUSE1I, andSEA-F(ab'), groups. The SEA-MUSE11 and SEA-F(ab')2 groups each consisted of 25mice (B and O. and the other groups consisted of 15 mice, respectively (A). *. P < 0.0001for SEA-MUSE11 Ab versus control group; and **. P < 0.01 for SEA F(ab')2 MUSEI 1Ab versus control group (both by Student's / test).

The activated T cells proliferate and produce cytokines, e.g., tumornecrosis factor a, IFN-y. IL-1. IL-2, and IL-6 (11-14). Consequently,the activated T cells rapidly bind to MHC class 11-expressing tumor

cells in the presence of SAgs and lyse them (15). To apply this activityto various tumor cells, regardless of the expression of MHC class IImolecules, SAg-conjugated mAb to tumor-associated antigens orcombined use of SAg-activated T cells with bispecific antibodies

directed to activated T cells and tumor cells has been introduced toexperimental immunotherapy (16-20). T cells activated with SAgs

have also been applied for in vivo experimental studies (21, 22).In the present study, we constructed SEA-MUSE 11 for retargeting

T-LAK cells to a BDC-grafted SCID mouse tumor. LAK therapy has

been used for clinical control but has hitherto not generated adequateresults (23, 24) because of an insufficiency in the targeting ability invivo. T-LAK cells do exert cytotoxicity against cancer cells if appro

priately targeted (2) and substantial quantities can be induced moreeasily than is the case with cytotoxic T cells. This is the reason whythey were used as effector cells in this study, with addition ofSEA-MUSE11 and SEA-F(ab')2 to improve the targeting. The spe

cific targeting ability against MUC1 antigen on the target cells wasenhanced with both SEA-MUSE11 and SEA-F(ab'), (Table 1 and

Figs. 1 and 3). Furthermore, SEA continued to stimulate T-LAK cells

(Fig. 2). However, administration of T-LAK cells preincubated with

SEA did not result in growth inhibition in the xenografted tumormodel. The experimental study only showed marked reduction oftumor growth in the groups using T-LAK cells with SEA-MUSE 11 orSEA-F(ab')2 (Fig. 6). These results indicate that both retargeting of

T-LAK cells by antibodies and activation with SEA are required for

effective immunotherapy in vivo. In an additional in vitro assay,T-LAK cells with a very low concentration of SEA-MUSE 11 (0.001

fig/ml) demonstrated significantly enhanced TN, supporting the efficacy of this approach in the xenografted tumor model.

The reason why preincubation of T-LAK cells with SEA for 6 hinduced enhanced TN index in the following 48-h MTS assay (Fig. 2),while the addition of SEA to T-LAK cells in the 48-h MTS itself was

without effect (Fig. 1) remains unclear. However, one possibility isthat a long incubation time, probably more than 48 h, might benecessary for further activation of T-LAK cells.

T-LAK cells infiltrating into tumors, demonstrated using polyclonalrabbit anti-human CD3 antibody, were more numerous after injectionwith SEA-MUSE11 than in the control groups (T-LAK alone, T-LAKinjected with SEA, and T-LAK injected with MUSE 11 Ab). This

correlated with results for adoptive immunotherapy in xenograftedmice.4

Marked reduction of tumor growth was observed for T-LAK cellswith SEA-MUSE11 or SEA-F(ab')2 in the tumor-xenografted model.

Although identical TN indexes were obtained for these two groups invitro (Fig. 1), the degree of tumor regression did vary slightly. It mightbe expected that SEA-F(ab'), would confer higher cytotoxicity and

tumor infiltration due to its smaller size than SEA-MUSE 11, but theaffinity difference between SEA-MUSE 11 and SEA-F(ab')2 (Table 1)

probably caused the difference in the in vivo results. To reduce theside effects of SAgs (25) by preventing SEA from binding to nontu-mor MHC class Il-positive cells, T-LAK cells preincubated withSEA-MUSE11 or SEA-F(ab')2 were administered to TFK-l-xe-

nografted SCID mice. For this purpose, use of mutated SEA (26),which has minimum class II binding, though maintaining the potentialfor T cell activation, should be considered for future study.

The reasons for selection of the MUC1 antigen as the target antigenin this study were its specificity and broad overexpression in variousepithelial cancers, such as those arising in the pancreas, stomach,ovaries, and bile ducts (27-29). Our previous study (2) demonstratedthat specific targeting therapy with T-LAK cells and two kinds ofbispecific antibodies (anti-CD3 X anti-MUCl and anti-CD28 X anti-

MUC1) inhibited the growth of xenografted human BDC. with thepercentage cytotoxicity found to be dependent on the degree of theMUC1 expression on the target cells. It has been reported that thebreast cancer-associated antigen, DF3/MUC1, deletes activated T

cells by induction of apoptosis and thus contributes to the paucity ofthe anti-breast cancer immune response (30). However, successfulinduction of anti-MUCl cytotoxic T cells has been reported (31).

Recently, we obtained killer cells with remarkable cytotoxicity bymixed lymphocyte tumor cell culture of peripheral blood mononuclearcells with autologous MUC1 vaccine cells.5 The mechanism of in

duction of killer cell apoptosis by MUC1 antigen is still controversial.With regard to this problem, administration of activated T-LAK cellsretargeted with SEA-MUSE 11 may be an effective approach to avoid

this inhibition of the antitumor immune response by MUC1.The injection of SAg into mice first induces strong T-cell activa

tion, followed by functional inactivation and deletion of T cells (32).

4 M. Shinoda and T. Kudo. Immunohistological study of adoptive immunotherapy with

SEA-conjugated antibody in xenografted SCID mice, manuscript in preparation.5 T. Kudo and M. Ichiyama. Induction of activated NK cells with strong cytotoxic

activity by MUC 1 vaccine cells, manuscript in preparation.

2842

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

ADOPTIVE IMMUNOTHERAPY WITH SEA-CONJUGATED ANTIBODY

It has been described that SEA-induced hyporesponsiveness involvesCD4+ cell deletion and a failure to produce cytokines in the remainingCD4+ T-cell compartment, whereas IFN-y production and cytotox-icity by the CD8+ T-cell compartment remain relatively intact (33).

However, IL-2 enhances and prolongs SEA-induced CTL activityafter injection of SEA, und exogenous IL-2 prevents the reduction in

cytotoxic activity after repeated injections with SEA (34, 35). Therefore, we consider IL-2 administration in combination with SEA-

conjugated Ab as essential to obtain the best advantage of SEA forsatisfactory antitumor effects. Although we used only T-LAK cells as

effector cells in this study, cytokine production from local T cells inthe tumor area by stimulation of SEA-MUSE 11 would be expected tocause synergistic effects on T-LAK cell activity in clinical treatment.

For application of SEA-conjugated Ab for clinical trials, single-

chain Fv antibody reconstituted from MUSE 11 Ab gene may beuseful because it lacks the Fc portion. Single-chain Fv antibody causes

only a low immunogenic response in the host due to its fast clearancefrom the circulation and therefore may not present a problem, evenafter repeated administration (36).

In conclusion, this is the first report of effective adoptive immu-notherapy by human T-LAK cells with SEA-conjugated antibodies inhuman s.c. tumor-grafted SCID mice. Although there are various

limitations, the results point to potential application in clinical therapyfor patients with BDC or other MUC1 -positive carcinomas.

ACKNOWLEDGMENTS

We thank Fujimi Koizumi for help in preparation of the manuscript.

REFERENCES

1. Suzuki. M., Takahashi, T.. Ouchi. K.. and Matsuno. S. The development and extension of hepalohilar bile duct carcinoma. A three-dimensional tumor mapping in theintrahepatic biliary tree visualized with the aid of a graphics computer system. Cancer(Phila.), 64: 658-666. 1989.

2. Katayose, Y., Kudo, T., Suzuki. M.. Shinoda. M., Saijyo. S.. Sakurai, N., Saeki. H.,Fukuhara. K., Imai. K., and Matsuno. S. MUC1-specific targeting immunotherapywith bispecific antibodies: inhibition of xenografted human bile duct carcinomagrowth. Cancer Res.. 56: 4205-4212, 1996.

3. White, J.. Herman, A., Pullen. A. M.. Kubo, R.. Kappler. J. W.. and Marrack, P. TheVß-specificsuperantigen staphylococcal enlerotoxin B: stimulation of mature T cellsand clonal deletion in neonatal mice. Cell, 56: 27-35. 1989.

4. Marrack, P.. and Kappler, J. The staphylococcal enterotoxins and their relatives.Science (Washington DC). 248: 705-711, 1990.

5. Hinoda. Y.. Arimura, Y., höh.F., Adachi, M., Tsujisaki, M.. Imai. K.. and Yachi. A.Primary structure of the variable regions of a monoclonal antibody MUSEI 1 recognizing the tandem repeat domain of a mucin core protein. MUCI. J. Clin. Lab. Anal..7: 100-104, 1993.

6. McKinney, M. M.. and Parkinson. A. A simple, non-chromatographic procedure topurify immunoglobulins from serum and ascites fluid. J. Immunol. Methods, 96:271-278, 1987.

7. Saijyo. S., Kudo, T., Suzuki. M.. Katayose, Y.. Shinoda, M., Muto. T.. Fukuhara, K..Suzuki. T.. and Matsuno. S. Establishment of a new extrahepatic bile duct carcinomacell line, TFK-1. Tohoku J. Exp. Med.. 177: 61-71, 1995.

8. Myers, D. E., Irvin, J. D., Smith, R. S., Kuebelbeck. V. M., and Uckun, F. M.Production of a pokeweed antiviral protein (PAP)-containing immunotoxin, B43-PAP. directed against the CD19 human B lineage lymphoid differentiation antigen inhighly purified form for human clinical trials. J. Immunol. Methods. 136: 221-238.

1991.9. Beun, G. D. M., Gorier, A., Nooyen, Y.. van de Yelde. C. J. H.. and Fleuren. G. J. T

cell retargeting using bispecific monoclonal antibodies in a rat colon carcinomamodel. II. Syngeneic colon carcinoma CC53I is efficiently killed by retargetedcytotoxic T lymphocytes in vitro despite limited lysis in "Cr release assays. J. Im

munol., 150: 2305-2315, 1993.

10. Haranaka. K.. Satomi, N.. and Sakurai. A. Antitumor activity of murine tumornecrosis factor (TNF) against transplanted murine tumors and heterotransplantedhuman tumors in nude mice. Int. J. Cancer. 34: 263-267, 1984.

11. Carlsson, R.. and Sjogren. H. O. Kinetics of IL-2 and interferon--y production,expression of IL-2 receptors, and cell proliferation in human mononuclear cellsexposed to slaphylococcal enterotoxin A. Cell. Immunol., 96: 175-183, 1985.

12. Fischer, H.. Dohlsten, M.. Andersson, U., Hedlund. G., Ericsson. P. O.. Hansson, J..and Sjogren, H. O. Production of TNF-a and TNF-ßby staphylococcal enterotoxin Aactivated human T cells. J. Immunol.. 144: 4663-4669. 1990.

13. Herrmann. T.. Baschieri. S., Lees, R. K.. and MacDonald. H. R. In vivo responses ofCD4* and CDS* cells to bacterial superantigens. Eur. J. Immunol., 22: 1935-1938,

1992.14. Dohlsten. M.. Sundstedt. A., Bjorklund, M., Hedlund, G., and Kailand. T. Superan-

tigen-induced cytokines suppress growth of human colon-carcinoma cells. Int. J.Cancer, 54: 482-488, 1993.

15. Dohlsten. M.. Lando. P. A.. Hedlund. G.. Trowsdale. J.. and Kalland, T. Targeting ofhuman cytotoxic T lymphocytes to MHC class II expressing cells by staphylococcalenleroloxins. Immunology. 71: 96-100. 1990.

16. Dohlsten. M.. Hedlund. G.. Akerblom. E.. Lando. P. A., and Kalland. T. Monoclonalantibody-targeted superantigens: a different class of anti-tumor agents. Proc. Nati.Acad. Sci. USA. X8: 9287-9291. 1991.

17. Dohlsten. M., Abrahamsen. L.. Björk.P.. Lando. P. A.. Hedlund. G.. Forsberg, G..Brodin. T.. Gascoigne. N. R. J., Förberg,C.. Lind, P.. and Kalland. T. Monoclonalantibody-superantigen fusion proteins: tumor-specific agents for T-cell-based tumortherapy. Proc. Nati. Acad. Sci. USA, 91: 8945-8949, 1994.

18. Thibault. C.. Nelson. H.. and Chapoval. A. I. Tumor-infiltrating lymphocytes can beactivated in situ by using in vivo activants plus F(ab')2 bispecific antibodies. Int. J.

Cancer. 67: 232-237. 19%.19. Penna. C.. Dean. P. A., and Nelson. H. Antitumor X anti-CD3 bifunctional antibodies

redirect T-cells activated in WTOwith staphylococcal enterotoxin B to neutralizepulmonary métastases.Cancer Res.. 54: 2738-2743, 1994.

20. Kuge. S.. Miura. Y.. Nakamura, Y., Mitomi, T., Habu. S.. and Nishimura. T.Superantigen-induced human CD4* helper/killer T cell phenomenon. Selective in

duction of Thl helper/killer T cells and application to tumor immunotherapy. J. Immunol., 154: 1777-1785, 1995.

21. Ochi. A.. Migita. K.. Xu. J.. and Siminovitch, K. In vim tumor immunotherapy by abacterial superanligen. J. Immunol.. 151: 3180-3186, 1993.

22. Lando. P. A.. Dohlsten. M.. Ohlsson. L., and Kalland. T. Tumor-reactive superantigens suppress tumor growth in humanized SCID mice. Int. J. Cancer. 62: 466-471,1995.

23. Rosenberg. S. A., Lotze, M. T.. Muul, L. M.. Leitman. S.. Chang. A. E.. Ettinghausen.S. E.. Matory. Y. L., Skibber, J. M.. Shiloni, E., Vetto. J. T.. Seipp, C. A.. Simpson.C.. and Reichert, C. M. Observations on the systemic administration of autologouslymphokine-activated killer cells and recombinant interleukin-2 to patients withmetaslatic cancer. N. Engl. J. Med., 313: 1485-1492. 1985.

24. Rosenberg, S. A.. Lotze, M. T.. Muul, L. M., Chang, A. E., Avis, F. P., Leitman, S.,Linehan. W. M.. Robertson. C. N.. Lee. R. E.. Rubin. J. T., Seipp. C. A.. Simpson.C. G., and White. D. E. A progress report on the treatment of 157 patients withadvanced cancer using lymphokine-activated killer cells and interleukin-2 or high-dose interleukin-2 alone. N. Engl. J. Med.. 316: 889-897, 1987.

25. Schlievert, P. M. Role of superantigens in human disease. J. Infect. Disease, 167:997-1002, 1993.

26. Hansson, J., Ohlsson, L., Persson. R.. Andersson, G., Ilback, N-G., Litton, M. J..

Kalland. T.. and Dohlsten. M. Genetically engineered superantigens as tolerableantitumor agents. Proc. Nail. Acad. Sci. USA. 94: 2489-2494, 1997.

27. loannides. C. G.. Fisk, B.. Jerome. K. R.. Irimura. T., Wharton. J. T.. and Finn, O. J.Cytoloxic T cells from ovarian malignant tumors can recognize polymorphic epithelial mucin core peptides. J. Immunol., 151: 3693-3703. 1993.

28. Ban. T.. Imai. K.. and Yachi, A. Immunohistological and immunochemical characterization of a novel pancreatic cancer-associated antigen MUSEI I. Cancer Res.. 49:7141-7146, 1989.

29. Yamashita. K.. Yonezawa, S.. Tanaka. S.. Shirahama, H., Sakoda, K.. Imai, K., Xing.P. X.. McKenzie, I. F.. Hilkens, J.. Kim, Y. S., and Sato. E. Immunohistochemicalstudy of mucin carbohydrates and core proteins in hepatolithiasis and cholangiocar-cinoma. Int. J. Cancer. 55: 82-91, 1993.

30. Gimmi. C. D.. Morrison, B. W.. Mainprice. B. A.. Gribben. J. G.. Boussiotis. V. A.,Freeman, G. J., Park. S. Y. L., Watanabe. M., Gong, J., Hayes. D. F., Kufe, D. W„and Nadler. L. M. Breast cancer-associated antigen. DF3/MUCI. induces apoptosis ofactivated human T cells. Nat. Med., 2: 1367-1370. 1996.

31. Takahashi, T.. Makiguchi. Y.. Hinoda. Y.. Kakiuchi. H.. Nakagawa. K.. Imai. K.. andYachi. A. Expression of MUCI on myeloma cells and induction of HLA-unrestrictedCTL against MUCI from a multiple myeloma patient. J. Immunol.. 153: 2102-2109.

1994.32. Kawabe. Y.. and Ochi, A. Programmed cell death and extrathymic reduction of V/38 *

CD4* T cells in mice tolerant to Slaphylococcus aureus enterotoxin B. Nature

(Lond.). 349: 245-248. 1991.33. Sundstedt, A.. Dohlsten, M., Hedlund, G., Hoidén,I.. Bjorklund, M., and Kalland. T.

Superantigens anergize cytokine production but not cytotoxicity in vivo. Immunology,82: 117-125. 1994.

34. Belfrage. H.. Dohlsten. M., Hedlund. G.. and Kalland. T. Enhanced and prolongedefficacy of superantigen-induced cytotoxic T lymphocyte activity by interleukin-2 inW'TO.Cancer Immunol. Immunother., 41: 87-94, 1995.

35. Belfrage. H.. Dohlsten. M.. Hedlund. G.. and Kalland. T. Prevention of superantigen-induced down-regulation of T-cell mediated cytotoxic activity by IL-2 in vivo.Immunology. 90: 183-188. 1997.

36. Kalinke. U., Krehber, A., Krebber. C.. Bucher. E., Plückthun,A.. Zinkemagel, R. M..and Hengartner. H. Monovalent single-chain Fv fragments and bivalent miniantibod-

ies bound to vesicular stomatitis virus protect against lethal infection. Eur. J. Immunol.. 26: 2801-2806, 1996.

2843

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

1998;58:2838-2843. Cancer Res   Masao Shinoda, Toshio Kudo, Masanori Suzuki, et al.   Xenografted Severe Combined Immunodeficient Micewith Bacterial Superantigen-conjugated Antibody to MUC1 in Effective Adoptive Immunotherapy by T-LAK Cells Retargeted

  Updated version

  http://cancerres.aacrjournals.org/content/58/13/2838

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

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

.http://cancerres.aacrjournals.org/content/58/13/2838To request permission to re-use all or part of this article, use this link

Research. on September 17, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from