[CANCERRESEARCH56. 1707-1712, April 15. 19961

Advances in Brief

Involvement of B Lymphocytes in the Growth Inhibition of Human Pulmonary

Melanoma Metastases in Athymic flu/flu Mice by an Antibody-Lymphotoxin

Fusion Protein1

Ralph A. Reisfeld,2 Stephen D. Gilies, John Mendelsohn, Nissi M. Varki, and JürgenC. BeckerDepartment of Immunology, The Scripps Research institute, La Jolla, California 92037 (R. A. R., J. C. B.J; Fuji imniunoPharmaceuticals Corporation, Lexington, Massachusetts02173 (S. D. G.J; Memorial Sloan-Kettering Cancer Center. New York. New York 10021 (J. MI; and Department of Pathology, School of Medicine, University of California,L.a Jolla, California 92037 [N. M. V.]

Abstract

Antibody.cytokinefusionproteinscan target biologicallyactivecytokinesto varioustumor sites,achievinglocalconcentrationssufficienttoinducehost immuneresponsesleadingto tumor elimination.Here, wedemonstratethe therapeuticefficacyof a tumor-specificantibody-lym.photoxinfusionprotein(ch225-LT)onxenograftedpulmonarymetastasesof humanmelanoma.In vitrostudiesindicateda directcytotoxiceffectofsuchconstructson melanomacellsvia the inductionof apoptosis,asdemonstrated by cell cycle analysis and DNA fragmentation. However,ch225-LTlackedany therapeuticeffectin immunedeficientC.B17scid/beige and scid/scid mice, indicating the insufficiency of this direct mechanism in vivo. In contrast, in athymic nu/nu mice, ch225-LT completelyinhibited outgrowth of the xenografted tumor. This therapeutic effect wasaccompanied by infiltrations of CD4S+, Mac-1+, and aslalo-GM1+ cellsinto the tumor; B220+ cellswerepresentin the surroundingtissueandtheperipheryof thetumor.The functionalroleof asialo-GM1+cellswasconfirmed by in vivo depletion studies. Our data indicate that an antibodylymphotoxinfusionproteineffectivelyinhibitsthegrowthofdisseminatedmelanomametastasesby mechanismsthat functionin the absenceofmatureT cells,but requireB, NK, and otherasialo-GM1+ cells.

Introduction

LT3 is closely related to TNF. LT is a Mr 25,000 glycoproteinproduced by mitogen-activated 1, B, and LAK cells (1, 2). LT and

TNF share two common receptors, p80 and p60, and have many

similar activities, including cytotoxic effects on some neoplastic cells

in vitro (1, 3). TNF secreted by tumor-specific i'lL cell clones isolatedfrom the primary and secondary melanoma of patients was reported tohave a direct cytotoxic effect against the relevant tumor (4). HumanLT was shown to enhance the susceptibility of preneoplastic andneoplastic cells to NK cell-mediated cytotoxicity (5). It exerts at least

equal antitumor activity in comparison to TNF and is less toxic inmice (6). Furthermore, only LT has a stimulatory effect on the growth

of prestimulated B cells (7) and acts as an autocrine growth factor forEBV.infected cells (8). These characteristics made TNF and LT

interesting candidates for tumor therapy. However, systemic applica

tion of effective doses of these cytokines has been prohibited by

severe toxicities of such treatments. In clinical application, high dosesof TNF could only be used regionally in the form of hyperthermicisolated limb perfusion, which proved to be an effective treatment of

Received 1/22/96; accepted 3/1/96.Thecostsof publicationof this articleweredefrayedin part by the paymentof page

charges.This article mustthereforebe herebymarkedadvertisementin accordancewith18U.S.C.Section1734solely to indicatethis fact.

I This work was supported in part by Outstanding Investigator Grant CA42508-10 (to

R. A. R.). J. C. B. is a fellow of the Deutsche Forschungs Gemeinschaft. This is TheScripps Research Institute Manuscript No. 9958-1MM.

2 To whom requests for reprints should be addressed, at Department of Immunology,

The ScrippsResearchInstitute, 10666North Torrey PinesRoad,La Jolla,CA 92037.3 The abbreviations used are: LT, lymphotoxin; TNF, tumor necrosis factor; mAb,

monoclonalantibody.

metastatic melanoma confined to one extremity. However, TNF leakage occurred occasionally, which caused major systemic effects, consisting of cardiovascular, respiratory, and hematological disturbances

(9).The ex vivo transfection of tumor cells with genes leading to the

expression of various cytokines offers a new approach for localized

immunotherapy (10, 11). The induction of inflammatory responses by

such cytokine-producing tumor cells led to their rejection upon inoculation into syngeneic mice and, in some cases, produced systemicimmune responses against challenges with wild-type parental tumorcells. T cells were shown to play a key role in all such experiments inwhich systemic immunity was analyzed. However, Qin and Blankenstein (12) recently found evidence for B lymphocyte involvement inantitumor responsesof syngeneic mice inoculated with murine plasmacytoma cells transfected with the human LT gene. This distinguishes LT from other cytokines analyzed in similar experiments.These investigators concluded that the LT-producing plasmacytomacells are rejected by a complex immunological mechanism involving

T cells, as well as B and other cells.We previouslydescribeda novel cytokine-basedimmunotherapeu

tic approach for cancer treatment that is characterized by a simplemodus operandi and effective local concentrations of cytokines in the

tumor microenvironment. We achieved this goal by constructingfusion proteins consisting of tumor-specific mAbs and cytokines, thus

utilizing the unique targeting ability of antibodies to direct cytokinesto tumor sites (13—15).Here, we demonstrate that an antibody-LTfusion protein can effectively inhibit growth of disseminated pulmonary metastasis of human melanoma in athymic nu/nu mice. Because

this therapy is not effective in either nude mice depleted of asialoGM1 + cells, C.B17 scidiscidmice lacking in B and T cells, or inC.B17 scidibeige mice that additionally are deficient in NK cells, theantitumor effects achieved by the ch225-LT fusion protein involve a

mechanism that functions in the absence of T cells, but requires B,NK, and other asialo-GM1 + cells.

Materials and Methods

Cell Linesand Reagents.The humanmelanomacell line M24met,cxpressing the EGF receptor, has been described previously (16). This cell linewas maintained as a monolayer in RPM! 1640 supplemented with 10% FCSand 2 mM L-glutamineand was passagedas necessary.The mouse/humanchimeric antibody directed against the human EGF receptor (ch225) wasconstructed by joining the cDNA for the variable region of the murine antibodywith the constant region ofthe yl heavy chain and the ic light chain, as reportedpreviously (17). The antibody-LT fusion protein (ch225-LT) was constructedby fusion of the LT cDNA to the yl H chain constant region at the end of thethird (CH3) domain, as described for a LT fusion protein with antigangliosideGD2mAb chl4.l8 (18).Thefusedgeneswereinsertedinto thevectorpdHL2encoding the ch225 V regions and the dihydrofolate reductase gene. Theresulting expression plasmid was introduced into Sp2/O-Agl4 cells by proto

1707

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

GROWThINHIBITIONOFMELANOMABYLI FUSIONPROTEIN

Immunohistology. Frozensectionswerefixed in cold acetonefor 10mmfollowed by removal ofendogenous peroxidase with 0.03% H202 and blockingof collagenouselementswith 10%species-specificserumin 1% BSAIPBS.The antibodieswerethenoverlaidonto serialsectionsat predetermineddilutions(usually20 @.&g/mi),andslideswereincubatedina humidchamberfor30mis. With PBS washesbetweenall steps,a biotinylated link antibodywasappliedfor 10mm,followed by another10mmexposureto eitherperoxidaseor alkalinephosphataselinked to streptavidin.The antibodiesusedwere ratIgO2b anti-CD45(clone 30F11.1), rat IgO2a anti-CD45R/220(cloneRA3—6B2),andratIgG2banti-Mac-l(cloneMl/70), allobtainedfromPharMingen(SanDiego,CA).

Results

Prior studies demonstratedthat fusion of a synthetic sequencecoding for human LT to the carboxyl end of the human Cyl gene ofmouse/humanchimericantigangliosideGD2 mAb chl4.18 resultedina fusionproteinthat retainedboth the antigenbindingcapacityof themAb and the cytokine activity of LT (18). A LT fusion proteinutilizing the antigen binding capacity of the mouse/human chimericantibody ch225 directed against the EGF receptor (ch225-LT) wasmade by the same procedure.Flow cytometry analysis of humanmelanoma cells M24met indicated that the ch225-LT fusion proteinrecognized EGF receptors expressed on this cell line equally well asits parent mAb ch225 (Fig. 1A). The functional activity of ch225-LTwas demonstrated by its cytotoxic effect against murine L-929 fibrosarcoma cells. A comparison of the cytotoxic effect exerted bych225-LT and rhLT against L-929 cells demonstrated that the fusionprotein was approximately 10-fold less active on a weight basis butonly 2-fold less active on a molar basis as measured by an l8.h51Cr-releaseassay (Fig. lB).

Becausemembersof theThE “family―werereportedto mediatetumor cell death (16, 17), we determined the direct cytotoxic effect ofthe ch225-LT fusion protein on M24met melanoma cells. As demonstrated by cell cycle analysis, both rhLT and ch225-LT effectivelyinduced apoptosis in these cells (Fig. 2, A-C). This fmding wasconfirmedby detectionof DNA fragmentationafter incubationwithboth rhLT and ch225-LT (Fig. 2D). The induction of apoptosisintumor cells in vivo may provide an effective mechanism to controltumor growth without relying on the host immune system. To test thishypothesis,we useda numberof inbredmousestrainsdisplayingdifferent immunodeficienciesfor injection with human melanomacells. In C.Bl7 scid/beige mice that are deficient in B, T, and NKcells, we found that neither rhLT nor the ch225-LT fusion protein hadany inhibitory effect on the growth of pulmonary melanomametestases(data not shown). Identical results were obtained when the sameexperimental protocol was used in C.Bl7 sckliscid mice that aredeficient in B and T but not NK cells (date not shown).

Becauseneither rhLT nor the ch225-LT fusion protein inducedantitumor effectors in C.B17 scid/scid or scid/beige mice in theabsenceof B, T, andNK cells, we addressedthe questionof whetherB cells,in theabsenceof T cellscouldmediateantitumoreffectsafterstimulation with rhLT or ch225-LT. We tested this hypothesis inathymicnu/numicethat lack T but not B cells.The dataof twoseparateexperiments shown in Table 1 suggest that B cells are indeedinvolved in the eradication of pulmonary melanoma metastases in

duced by the ch225-LT fusion protein. It is evident from theseexperimentsthatch225-LTwasquiteeffectivein thisregard,whereasrhLT was ineffective at the dose level tested. The antitumor effectinduced by ch225-LT is dose-dependent,as it is lesspronounced at the4-;Lg than the l6-@.agdose level (Table 1, Experiment 1).

Becauseathymic nu/nu mice possessboth B and NK cells, weaddressedthe question of whether B cells alone were involved in theantitumor effects induced by ch225-LT. To this end, we depleted

plastfusionandselectedin DMEM supplementedwith 10%fetalbovineserumand 100 flM methotrexate.The fusion protein was purified over a proteinA-Sepharoseaffinity column,asdescribedpreviously(18).All otherantibodies used are commercially available and have been described by their respecfive manufacturers(PharMingen,SanDiego,CA, andWAKO Bioproducts,Richmond, WA).

Animals. C.Bl7 scidiscid and athymic nu/nu mice were obtained fromTacomcFarmsat the ageof 6 weeks.C.Bl7 scidlbeigemice wereobtainedfrom The JacksonLaboratoryat the ageof 4—6weeks.Theseanimalswerehousedunder specific pathogen-freeconditions and all experimentswereperformedaccordingto NIH guidelinesfor careanduseof laboratoryanimals.

in Vito Depletion with mAb. Rabbit anti-asialo-GM1 antiserum (WAKOBioproducts)wasusedfor in vivo depletionexperiments.Protocolsleadingtomaximumdepletion,asdeterminedby indirect immunofluorescencestainingandcytofluorometricanalysis,consistedof injectionsof 100 @.dof antiserumon days —3,0, 7, 14, and 21.

Tumor MOdels.Experimentallung metastaseswere inducedby injectionof singlecellsuspensionsof 1 X l0@M24metcellsintothelateraltailvein.Toprevent pulmonary embolism caused by injection of tumor cells, mice wereanesthetizedby halothaneinhalation,andtumorcellsweresuspendedin 500 @dPBScontaining0.1%BSAandadministeredi.v. overa periodof 60 s.At day35 after tumor cell injection, grosslyvisible metastaseswerepresenton thesurfaceof theorgan.Lungswerefixed in Bouin fixative andexaminedundera low-magnificationmicroscopefor tumorfoci on their surface.SectionsfromthelungswerestainedwithH&E andexaminedhistologically.Treatmentofanimals was started 24 h after induction of metastases and consisted of i.v.injection of either 0.5 @tgLT, 4 @&gch225-LT, 16 @.&gch225-LT, or PBS for 7days. Animals were sacrificed 28 days after tumor cell inoculation.

S.c. tumorswere inducedby s.c. injection of 2.5 X l0@M24metcells inRPMI 1640,which resultedin tumorsof approximately50 p1volumewithin14days.Tumorsandsurroundingtissueswereexcisedatindicatedtimepointsandsnapfrozen.

Flow Cytometry.Cells (1 X l0@)were washedtwice with PBS andincubatedfor 30 mm with the primary mAb ch225or the ch225-LT fusionprotein. After two washes with HBSS containing 0.2% sodium snide, cellswere stainedfor 30 mm with a mouseantihumanIgG FITC conjugate.Allincubations and washes were performed at 4°C.Cells were analyzed with aFACScan(BectonDickinson,SanJose,CA) at488 am.

Cytotoxicity Assay. Cell lysis was measured in an 18-h 51Cr-releaseassay.U-bouomedplateswereusedfor cytotoxicityassaysin conjunctionwith RPMI1640 supplemented with 10% FCS. Assays were performed at various rLT andch225-LT concentrations using 2.5 X iO@“Cr@labeledL-929 murine sarcomacells/well. Plates were incubated for 18 h, then a 70-p.l aliquot was removedfrom eachwell, and the specific “Cr-releasewas measured.Resultswereexpressed as:

% cytotoxicity =

experimental release (cpm— spontaneous release (cpm)@

maximum release (cpm) —spontaneous release (cpm)

where spontaneous release was assessed by incubating target cells in theabsenceof LT, andmaximumreleasewasdeterminedin the presenceof 1%Triton X-lO0.

Cell Cycle Analysis. Cell cycle analysisof M24metmelanomacells wasperformed by measuring nuclear DNA content after incubation for 24 h at37°Cwith eithermAb ch225,rhLT, or ch225-LTfusionprotein.Briefly,M24met cells were fixed in ice-cold 70% ethanol overnight at 4°Cand stainedfor 30 mmatroomtemperaturewithpropidiumiodide(50 @ag/m1)containing100 Kunitz units of RNase A. Cells were analyzed by flow cytometry in afluorescence-activatedcell sorter.

Isolation of Low-Molecular-WeightDNA. DNA fragmentationwasmeasured by isolation oflow-molecular-weight DNA from M24met cells after

treatmentwith eithermAb ch225,diLl, or ch225-LT.Briefly, cellswerelysedin 0.1 M Tris-HC1 buffer (pH 7.4), 10 mr@tEDTA, and 0.2% Triton X-l00.After centrifugationat 14,000g, DNA wasextractedfrom thesupernatantwithphenol/chloroform. After ethanol precipitation, the DNA was resuspended in

Tris-EDTA buffer (pH 7.4)andincubatedwith 100Kunitz unitsof RNasefor60 mm at 37°Cbeforeelectrophoreticanalysison a 2%agarosegel for 45 mmat 80 mV.

1708

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

GROWTh INHIBITION OF MELANOMA BY LI FUSION PROTEIN

(PBS)andmice injectedwith rhLT revealedalmostno infiltrateseither within or at the periphery of the tumor, with only occasionalCD45+ and asialo-GM1 + cells (data not shown). In contrast, tumorsof mice treated with the ch225-LT fusion protein showed massiveinfiltration ofCD45+ (Fig. 3A), Mac-i + (Fig. 3B) andasialo-GM1+cells (Fig. 3C). Furthermore,B220+ cells could be detectedat thetumor site;however,thesecellswererestrictedto theperipheryof thetumor and the surroundingtissues(Fig. 3D).

Discussion

We recently reported that antibody-cytokinefusion proteins caneffectively targetcytokinesto tumor sites,achievinglocal concentrations sufficient to induce eradication of disseminated metastases(14,15). We demonstrate here the therapeutic efficacy of an anti-EGF

receptor antibody-LT fusion protein (ch225-LT) in inhibiting thegrowth of disseminated pulmonary metastasesof human melanoma inathymic (nu/nu) mice. The antitumor effect of ch225-LT involves a

complex immunological mechanism that functions in the absenceofTcells but requires B cells and asialo-GM1 + cells, which are mainlyNK cells. This conclusion is supported by the fact that growth ofdisseminated pulmonary melanoma metastasesin C.B17 scid/scid orscid/beige mice could not be inhibited by the ch225-LT fusion protein.However, suchmetastasescould be effectively eradicatedby ch225-LT, but not rhLT, in athymicnu/nu mice. The requirementfor NKcells in these animals was evident from depletion studies with antiasialo-GM1 antibody before and during treatmentwith ch225-LT,resultingin completeabrogationof its antitumoreffect. However, itshould be noted that asialo-GM1 antiserum also reacts with non-NKcells, such asmonocytes, bone marrow, spleencells, and macrophagesof nude mice. It is possible that the differences in efficacy ofch225-LT in C.B17 scid/scid and athymic nu/nu mice may also be

influenced by differences in NK cell activities among these animals.In addition, antihuman B cell responses and production of 1gM in

athymic nu/nu mice bearing human M24met xenografts could alsocontribute to the therapeutic effect of ch225-LT, as such antibodies

and NK cells might kill tumor cells by antibody-dependent cellularcytotoxicity.

Similar to our findings, Funahashiet aL (19) reported that the@1 antitumor effect of LT was abrogated by anti-asialo-GM1 antibody in

I@ asyngeneicMethAfibrosarcomamodel.Ourfindingsagree,inpart,with those of a recent report by Qin and Blankenstein(12). Theseinvestigators used murine plasmacytoma J588L cells, transfected witha human LT expression plasmid and obtained an increased localconcentration of this cytokine. They also found evidence for B lym

phocyte involvement in athymic nu/nu mice, in which suppression ofs.c. tumor growth was considerably more effective than in C.Bl7

scid/scid mice (12). It should be noted that a defect in B cell development of nude mice was suggestedby Wortis et a!. (20) based,inpart, on a difference in electrophoretic mobility from that of normal Bcells and the inability of added T cells to release nude mouse B cellsafter binding to macrophages; however, it is unlikely that this defect

has an impact on our observations.Immunological analysis confirmed the involvement of B and NK

cells in the ch225-LT-mediated antitumor effect. Interestingly, asialoGM1 + cells were found to be infiltrating the whole tumor, whereasB220+ cells were restricted to the tumor periphery. Qin and Blankenstein (12) demonstrated the presence of B cells throughout the

entire tumor, provided the tumor cells were producing sufficientamountsof LT. Thesedifferencescanbeexplainedby the concentration gradientsof LT in the tumor microenvironmentcreatedby thesedifferent forms of therapy,becauseantibody-targetedcytokinedclivcry produces higher concentrations in the periphery.

A7OO@

560

.! 320:C

(@ 280:

I 40:

B100'

@25'

c7

i6@1@

1709

: :: : Il!tI: : I': : It: : it: : I, I: : &: :@: :@: :•: :,:@ I: I: ‘, I

U I I lull ‘—r—1—r1@;;;@——t—@ lull I I I I 11111

100 101 1O@ 1O@ 1O@Fluorescence

pg proteInFig. 1. Biological activities of the ch225-LT fusion protein. A, flow cytometiy of

1 X l0@M24met cells incubated 30 mm with PBS (. ) ch225 (- ), and ch225-LTfusion protein (—). B, cytotoxicity assay of 2.5 X l0@'51Cr-labeled L-929 murinesarcoma cells with various concentrations of rhLT (0) and ch225-LT fusion protein (0).

asialo-GM1+ cells in vivo with a polyclonal rabbit antibodyandthentested the effect of ch225-LT against M24met pulmonary metastasesin these animals. The data depicted in Table 2 indicate that themechanism involved in the antitumor effect induced by ch225-LT israther complex and requiresnot only B cells, but also a variety ofasialo-GM1 + cells. Further information on the cellular mechanism

involved was provided by immunohistochemicalanalysisof established s.c. tumors. We demonstratedpreviously that antibody-cytokine fusion proteins accumulate in s.c. tumors expressing the respective target antigen (1 1, 12). In this study, four animals each weretreated daily over a period of7 days by i.v. administered PBS (0.2 ml),rhLT (0.5 ILg), or ch225-LT fusion protein (16 @g)starting 10 daysafter tumor cell inoculation. Biopsies were taken on days 4, 7, and 10after the start of therapy. Inflammatory infiltrates were very obviouson day 7. Biopsiesfrom tumors(-@-50@.dvolume) of controlanimals

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

A361@@

200 400 600B361

@ Mj@AM20 2@O46O@6oCD

M 1 234

‘@9. -—36@.M2

M1@ft,40

200 400 600

TabIc I Therapeuticefficacyof ch225-LT on experimental lungmeta.ctases of humanmelanomainathymic nu/numiceTreatment―No.

offocibExperiment

I'PBS>500>500>500>500>500>500rhLT

(0.5 gLg)ch225-LT (4 1Lg)@ch225-LT(l6@gY'>500

12>500 14>500 26>500 100o>500

>5002>500

>5008Experiment

2@PBS>500>500>500>500>500>500>500rhLT

(0.5@@g)>500>500>500>500>500>500>500>500ch225-LT(I6@g)000002821

GROWTH INHIBITION OF MELANOMA BY LT FUSION PROTEIN

Fig. 2. Determination of cell death by apoptosis. Flow cytometry of I X l0@M24met cellsincubated for 24 h with 100 p@g/m1ch225 (A ), 0.15gag/mirhLT (B), and 3 g@g/mlch225-LT (C). D,DNA fragmentation in M24met melanoma cells:M, standard; Lane I, M24met control; Lane 2,M24met + ch225 (100 @ag/ml);Lane 3,M24met + ch225-LT fusion protein (5 @sg/ml);Lane 4, M24met + rhLT (0. 15 g.tg/ml).

The effect of LT on B cells in our tumor model system is inaccordancewith previously reported in vitro data. LT was shownto be a growth factor for prestimulatedhuman B cells (2 1) and wasreported to act as an autocrine growth factor for human B cell lines(8) and for EBV-transformed B cells (10). Interactions between Band NK cells are suggestedby a report that NK-mediated regulation of human B cells involves members of the TNF family ofcytokines (22). A possible interaction between murine B cells andhuman LT is suggested by the finding that Hodgkin's diseasederived cell lines secreting large amounts of LI (23) grow astumors in C.B 17 scid/scid, but not in athymic nu/nu mice (24). In

addition, based on published data from several studies, LI caninitiate growth stimulation of B cells (5) and the Mr 75,000TNF/LT receptor molecule was shown to be functionally active

when expressed on B cells (25). Activated B cells, in turn, areknown to produce ILl2, which induces the production of IFN-yfrom NK cells, thus activating them and enhancing their cytotoxicactivity against tumor target cells (26). Hypothetically, this is a

mechanism that may be involved in the eradication of pulmonarymelanoma metastasesdescribed in this study.

In summary, we have demonstratedthat an antibody-LI fusionprotein targets sufficient LI to disseminated pulmonary melanoma

a Treatment was started 24 h after inoculation of tumor cells and consisted of iv. injection of either PBS (0.1 ml). rhLT (0.5 pg), or ch225-LT at either 4 gagor 16 g.@gdaily over7 consecutive days, as indicated. Animals were sacrificed 28 days after tumor cell inoculation.

b All experimental groups were started with eight mice. Animals found dead before the planned date of sacrifice were not included in the evaluation.

C Experimental lung metastases were induced by iv. injection of I X 106 M24met melanoma cells into athymic nu/nu mice.

d Differences in the number of metastatic foci between the groups of mice receiving either 4 @gor 16 @gch225-LT daily for 7 days were statistically significant (P@ 0.002), as

determined by paired Student's t test.

Table2 Effectof ch225-LTonexperimentallungmetastasesof humanmelanomain athymicpiulmumicedepletedofasialo-GMJ+ cells

No. of FocibTreatment'@

PBS>500>500>500>500>500>500rhLT>500>500>500>500>500>500ch225-LT'0000034ch225-LT

plus ani1 d681 00>500>500>500>500

a Treatment was started 24 h after inoculation of tumor cells and consisted of iv. injection of either PBS (0.1 ml), rhLT (0.5 @xg)or ch225-LT (l6@xg),dailyover 7 consecutivedays. as indicated.

b Experimental lung metastases were induced by intravenous injection of I X 106 M24met melanoma cells into groups of six athymic nu/nu mice.

C Differences in the number of metastatic foci between the groups of mice receiving the fusion protein, either depleted or not depleted of asialo-GM 1 + cells, were statistically

significant (p 0.002), as determined by paired Student's t test.d Mice were depleted of asialo-GMI + cells by injection of anti-asialo-GMI antiserum (100 @d) before and after tumor cell inoculation on days —3, 0, 7, 14, and 21.

1710

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

GROWTH INHIBITION OF MELANOMA BY LI FUSION PROTEIN

B “‘ F .@..:. i::.' .‘@ .@ .@ .;.

,@@@ ?,

@,4: @: :,@@ ‘@•‘... .;e,•.''_ .@.I@..•;.,;..‘:‘@‘@.@@@ ‘@ ‘@:y..@ ‘@@ f?@@i:L@'tc!@―t@@@ ••@ : @‘

/. # e@@s@1@; ..@

@s@@ ?@W@r@ I@@@@

. . ,.@ ,\ ‘@j@@@ •@@

,@ ‘—@ .@

. , , . . ,@ :@ . .@

;@ :@‘@@@ .@..

‘. ,‘ 4

. ‘ @. .--.a@.. •@@

Fig. 3. Immunohistologicalcharacterizationof inflammatoryinfiltratesinducedbych225-LTfusionproteinin M24mettumors.Athymic nulnumicewereinjecteds.c.with 2.5 X l0@M24met cells. After 10 days, four animalseach were injected iv. with 16 @xgch225-LT/day for 7 days. On day 7, after the startof therapy,sectionsof thesetumorswere preparedand subjectedto anti-CD45 (A), Mac-l (B), anti-asialo-GMI (C), and anti-B220 (D) followed by immunostainingand counterstainingwith hematoxylin. X200.

, . .- . . . .

I' ‘ :

, . . ,,@‘, , . . .

-c :,.,@—•@ ,;@. , ,. . ,.

.\ 4 @;:‘ -‘:@@ ::@@@

4@ t :@@ t@;@@ J@@ @) :@@@

/@, ,. @4@' 4 ‘@. @s

.. ,@@ :@ ‘.@ r-@ ,.•-‘ : ‘, ,@@@

; x@ j@@ I@ .,t. -,@ —-i@ t

A@

metastasesto eradicate them in athymic nu/nu mice, but not in C.B I 7scid/scid or scid/beige mice. A complex immunological mechanismappearsto be involved in this antitumoreffect that functionsin theabsenceof mature I cells, but requires B, NK, and other asialoGM1+ cells.

References

I . Chaturvedi, M. M., LaPushin, R., and Aggarwal, B. B. Tumor necrosisfactor andlymphotoxin.J. Biol. Chem., 269: 14575—14583,1994.

2. Aggarwal,B. B. Comparativeanalysisof the structureand functionof TNF-a andTNF-13. In: B. B. Aggarwal and J. Vilcek (eds.), Tumor Necrosis Factors: Structure,Functionand Mechanismof Action, pp. 61—78.New York: Marcel Dekker, Inc.,I 992.

3. Desch,C. E.. Dobrina, A., Aggarwal, B. B., and Harlan, J. M. Tumor necrosisfactor-a exhibits greater proinflammatory activity than lymphotoxin in vitro. Blood,75: 2030—2034, 1990.

4. Becker,J.C., Schwinn,A., Dummer,R.,Burg,G., andBrocker,E-B. Lesion-specificactivationof clonedhumantumor-infiltratinglymphocytesby autologoustumorcells:inductionof proliferationand cytokine production.J. Invest. Dermatol., 101: 15—21,1993.

5. Reiter, Z., and Rappaport.B. Dual effects of cytokines in regulationof MHCunrestrictedcell-mediatedcytotoxicity. Crit. Rev. Immunol.,13: 1—34,1993.

6. Qin, Z., vanTits, L. J. H., Buurman,W. A., andBlankenstein,I. Human lymphotoxinhasat leastequal antitumoractivity in comparisonto human tumor necrosisfactor.but is lesstoxic in mice. Blood,85: 2779—2785,1995.

7. Abken,H., Fluck,J.,andWillecke,K. Fourcell-secretedcytokinesactsynergisticallyto maintain long-termproliferationof humanB cell lines in vitro. J. Immunol., 149:2785—2794,1992.

8. Gibbons,D. L., Rowe,M., Cope,A. P., Feldmann,M., and Brennan,F. M. Lymphotoxinactsasan autocrinegrowthfactorfor Epstein-Barrvirus-transformedB cellsand differentiated Burkitt lymphoma cell lines. Eur. J. Immunol., 24: 1879—1885,1994.

9. Lejeune,F. J., Lienard,D.. LeyVraz,S.,andMirimanoff, R. 0. Regionaltherapyofmelanoma. Eur. J. Cancer., 29A: 606—612,1993.

10. Colombo, M. 0., and Forni, G. Cytokine genetransferin tumor inhibition andtumortherapy: where are we now? Immunol. Today, 15: 48—51,1994.

11. Schmidt-Wolf, G., and Schmidt-Wolf, I. G. H. Cytokines and clinical gene therapy.Eur. J. Immunol., 25: 1137—1140, 1995.

12. Qin, Z., and Blankenstein, T. Tumor growth inhibition mediated by lymphotoxin:evidenceof B lymphocyteinvolvement in the antitumorresponse.Cancer Res., 55:4747—4751,1995.

13. Gillies, S. D., Reilly, E. B., Lo, K-M., and Reisfeld, R. A. Antibody-targetedinterleukin2 stimulatesT cell killing of autologoustumorcells.Proc.Nail. Acad. Sci.USA, 89: 1428—1432,1992.

14. Sabzevari,H., Gillies, S. D., Mueller, B. M., Pancook,J. D., and Reisfeld, R. A. Arecombinantantibody-IL2 fusion protein suppressesgrowth of hepatichuman neuroblastomametastasesin severecombinedimmunodeficiencymice.Proc.Natl. Acad.Sci. USA., 91: 9626—9630,1994.

15. Becker, J. C., Pancook, J. D., Gillies, S. D., Mendelsohn, J., and Reisfeld, R. A.Eradication of human hepatic and pulmonary melanoma metastasesin SCID miceby antibody-interleukin 2 fusion proteins. Proc. Natl. Acad. Sci. USA, in press,1996.

16. Mueller, B. M., Romerdahl,C. A., Gillies, S. D., and Reisfeld,R. A. Enhancementofantibody-dependentcytotoxicity with a chimeric anti-GD2 antibody.J. Immunol.,144:1382—1386,1990.

17. Gillies, S. D., La, K-M., and wesolowski, J. High-level expressionof chimericantibodiesusingadaptedcDNA variableregioncassettes.J. Immunol.Methods,125:191—202,1989.

18. Gillies, S. D., Young, D., Lo, K-M., Foley, S. F., and Reisfeld, R. A. Expressionofgeneticallyengineeredimmunoconjugatesof lymphotoxinanda chimericanti-gsagliosideGD2 antibody.Hybridoma.3: 347—356,1991.

19. Funahashi,H., Watanabe, H., Abo, T., Indo, K., and Miyaju, H. Tumor growthinhibition in mice by glycosylatedrecombinanthuman lymphotoxin: analysisoftumor-regional mononuclear cells involved with its action. Br. J. Cancer, 67: 447—455, 1993.

20. wortis, H. H., Burkly, L., Hughes,D., Roschelle,S.,andWaneck,G. Lackof matureB cells in nudemicewith x-linked immunedeficiency.J. Exp.Med., /55: 903—913,1982.

21. Kehrl, H. H., Alvarez-Mon.M., Delsing,G. A., andFauci,A. S. Lymphotoxinis animportant T cell-derived growth factor for human B cells. Science (Washington DC).238: 1144—1146,1987.

1711

@..:@ .:@@ .,‘

I,J@ ,:@l@,@

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

GROWTHINHIBITION OF MELANOMA BY LI FUSION PROTEIN

22. Becker,J. C., Kolanus,W., Lonnemann,C., and Schmidt,R. E. Human naturalkillerclones enhance in vitro antibody production by tumor necrosis factor a and yinterferon.Scand.J. Immunol.,32: 153—162,1990.

23.Foss,H.0., Herbst,H.,Oelmann,E.,Samol,J.,Blankenstein,T.,Matthes,J.,Qin,z., Falini, B., Piled, S., Diamantstein,T., and Stein, H. Lymphotoxintumornecrosis factor and interleukin 6 gene transcripts are present in Hodgkin andReed-Sternberg cells of most Hodgkin's disease cases. Br. J. Haematol., 84:627—635,1993.

24. von Kalle, C., Wolf, J., Becker, A., Sckaer, A., Munck, M., Engert, A., Kapp, U.,

Fonatsch, C., Komitowski, 0., Feaux de Lacroix, W., and DieM, V. Growth ofHOdgkin cell lines in severely combined immunodeficient mice. mt. J. Cancer, 52:887—891,1992.

25. Heilig, B., Mapara, M., Brockhaus,M., Krauth, K., andDorken, B. Two typesof TNFreceptors are expressed on human normal and malignant B lymphocytes. Clin.Immunol. Immunopathol.,61: 260—267,1991.

26. Trinchieri, 0. Interleukin-12. A proinflammatory cytokine with immunoregulatoryfunctionsthatbridgemateresistanceandantigen-specificadaptiveimmunity.Annu.Rev. Immunol., 13: 257—276,1995.

1712

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

1996;56:1707-1712. Cancer Res   Ralph A. Reisfeld, Stephen D. Gillies, John Mendelsohn, et al.   Mice by an Antibody-Lymphotoxin Fusion Protein

nu/nuHuman Pulmonary Melanoma Metastases in Athymic Involvement of B Lymphocytes in the Growth Inhibition of

  Updated version

  http://cancerres.aacrjournals.org/content/56/8/1707

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/56/8/1707To request permission to re-use all or part of this article, use this link

Research. on January 19, 2021. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from