Cytolysis of Tumor Cells Expressing the Neu/erbB-2, erbB-3 ...Vo!. 2, 1001-1008, June 1996 Clinical Cancer Research 1001 Cytolysis of Tumor Cells Expressing the Neu/erbB-2, erbB-3,

Vo!. 2, 1001-1008, June 1996 Clinical Cancer Research 1001

Cytolysis of Tumor Cells Expressing the Neu/erbB-2, erbB-3, and

erbB-4 Receptors by Genetically Targeted Naive T Lymphocytes

Uwe Altenschmidt, Raif Kahl, Dirk Moritz,

Barbara S. Schnierle, Bernhard Gerstmayer,

Winfried Wels, and Bernd Groner’

Institute for Experimental Cancer Research, Tumor Biology Center,

Breisacher Strasse 177. D-79106 Freiburg, Germany

ABSTRACT

We are developing strategies to use naive T lympho-

cytes in cancer therapy. For this purpose, we are deriving T

cells with specificity of recognition for defined tumor cells.

To direct effector lymphocytes toward tumor cells, we have

manipulated the recognition specificity of naive rat and

mouse T lymphocytes and a mouse T-cell line. The cells were

stably transduced with a chimeric T-cell receptor (TCR)

component. The � chain of the TCR consists of a single

transmembrane protein with a short extracellular domain

and an intracellular domain for TCR signaling. We pro-

vided an extracellular tumor cell recognition domain to the

� chain. Human heregulin f31 (ligand to the erbB-3 and

erbB-4 receptors) and three different single-chain antibodiesspecific for the human and rat Neu/erbB-2 receptors were

used. One single-chain antibody (Cli) is directed against the

rat Neu protein, and one single-chain antibody (FRP5) is

directed against the human erbB-2 receptor. The single-

chain antibody (R-AK) directed against the Mr 14,000 fusion

protein of orthopox viruses served as a control. An efficient

procedure was devised to introduce the chimeric genes intoprimary rat and mouse T lymphocytes. Retrovirus-produc-

ing packaging cell lines were cocultured with the T cells

activated by phytohemagglutinin and interleukin 2. T-cell

lines were transduced by exposure to retrovirus-containing

supernatants from helper cell lines. Expression of the fusion

genes was determined by fluorescence-activated cell sorting

analysis More than 80% of the naive rat and mouse T cells

and 85-100% of the cells from the established T-cell lines

expressed the fusion genes within 48 h after infection. The

expression of the fusion genes was maintained for at least 10

days after infection. Target cells expressing Neu/erbB-2,

erbB-3, or erbB-4 were lysed in vitro with high specificity by

T cells expressing the corresponding recognition proteins.No selection of a marker gene is necessary to confer a

predetermined recognition specificity. The described exper-

iments are important for a gene therapy approach to cancer

treatment with autologous T cells.

INTRODUCTION

T lymphocytes are the main effector cells operative in the

recognition and elimination of syngeneic tumors ( I ). Different

tumor cells have been found to possess tumor-specific antigens,

which can be recognized by T cells (2-4). T cells are able to

penetrate the core of solid tumors and to secrete cytokines,

which activate nonspecific effector cells. The transfer of T

lymphocytes with antitumor activity into cancer patients is one

approach of adoptive immunotherapy (5-7). Tumor-infiltrating

lymphocytes, in vitro sensitized lymphocytes derived from

CTLs (8), and lymphokine-activated killer cells (9) have been

shown to possess the potential to mediate tumor regression.

In clinical trials, however, it was shown that only a fraction

of the treated patients responded to the cell transmissions (10).

The efficiency of lymphocyte-mediated tumor therapy can po-

tentially be improved by in vitro manipulation of the cells. We

are developing methods of genetic manipulation of T-cell rec-

ognition specificity to target them toward tumor cells. For this

purpose, chimeric components of the TCR2 are derived and

expressed in transduced T cells.

The TCR is a multimeric complex composed of six differ-

ent subunits, TCR-a, TCR43, CD3-y, CD3& CD3#{128},and the

CD3� chain or its alternative splice product ‘q. The TCR-a43

heterodimers recognize antigen peptides in a complex with

MHC molecules on antigen-presenting cells ( 1 1 ). Because the

TCR-a/� proteins have only marginal intracellular domains,

CD3�(1) homodimers serve as transducers to the intracellular

signaling machinery. They play a central role in the induction of

cytolysis ( I 2-15).

Recently, it was demonstrated that fusion of extracellular

ligand-binding structures with the � chain results in molecules

that can be activated by interaction with ligands ( 16-19). T

cells, which express a single-chain antibody directed against the

human erbB-2 receptor, coupled with the � chain of the TCR,

conferred cytotoxicity in vitro and in vito in a MHC-independ-

ent manner ( I 7). The target structure, the erbB-2 receptor, is

overexpressed in about 30% of human breast and ovarian car-

cinomas (20-22). erbB-2, also called Her-2, belongs to the EGF

receptor family. Members of this family comprise the EGF

receptor (EGFRIerbB- I ). erbB-2/Her-2/neu, erbB-3/Her-3, and

erbB-4/Her-4 (for reviews, see Refs. 23 and 24). In addition to

the erbB-2 protein, the erbB-3 receptor is also overexpressed in

mammary tumors and tumor cell lines (25, 26). Whereas no

specific ligand has yet been found for the erB-2 receptor, the

Received I 1/1/95: revised 1/1 7/96; accepted 3/5/96.

I To whom requests for reprints should be addressed. Phone: (49)

761-206- 1600; Fax: (49) 761-206-1699.

2 The abbreviations used are: TCR, T-cell receptor; EGF, epidermal

growth factor; mAb, monoclonal antibody; scFv, single-chain antibody;

IL-2. interleukin 2; FACS, fluorescence-activated cell sorting; PHA,

phytohemagglutinin; VH, heavy-chain variable domain; V1. light-chain

variable domain; myc-tag, human c-invc epitope.

Research. on April 14, 2020. © 1996 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

1002 T-Cell-mediated Tumor Cell Lysis

erbB-3 and erbB-4 receptors are specifically activated by the

heregulins (27, 28).

The principle of MHC-independent tumor cell lysis with

genetically manipulated CTLs has been established by several

laboratories (16-19). We are concerned with the requirements

for the practical application of this strategy to cancer treatment.

To improve and expand this strategy, we describe the use of

primary T cells, and we present the construction, transduction,

and expression of four � chain fusion molecules. These mole-

cules were transduced into naive T cells and conferred the

MHC-unrestricted tumor cell lysis in vitro. We used as recog-

nition domains either the natural ligand to erbB-3 and erbB-4,

human heregulin [3 1 , or scFvs specific for the rat Neu protein,

specific for human erbB-2, or specific for a viral surface protein.

These binding structures were linked to a myc-tag recognition

epitope and the � chain of TCR-CD3 complexes. The immuno-

globulin-like hinge region of the CD8a domain (29) was in-

serted as a flexible spacer. � fusion genes were transduced in

naive rat and mouse T cells and a mouse CTL line. We inves-

tigated the conditions that yielded the highest transduction ef-

ficiency of the naive cells. A procedure was established with

which 85-100% of target cells could be transduced. Transduced

T cells expressed the fusion gene products, and in vitro, these

cells efficiently lysed syngeneic rat tumor cells expressing the

Neu protein or mouse fibroblasts expressing erbB-2, erbB-3,

and erbB-4. Our studies show that the majority of naive T

lymphocytes can be genetically manipulated without further

selection to gain new target cell recognition specificity.

MATERIALS AND METHODS

Cell Lines and Culture Conditions. Cl96 is a C57BL/6

mouse (H-2�’)-derived CTL line with H-2K�’-restricted specific-

ity for P815 (H-2”) mastocytoma cells (30). C196 cells were

maintained in DMEM (Life Technologies, Inc., Eggenstein,

Germany) with 10% FCS (Life Technologies, Eggenstein, Ger-

many) and 3% cell-free supernatant of murine IL-2-secreting

X63Ag8-653 cells (31). The retroviral packaging cell lines ��E

(32) and PA317 (33), the fibroblast cell line NIH3T3, and the

BDIX rat neuronoma cell line NV2Cd3 were cultured in DMEM

with 10% FCS. HC1 lRl 1 1 cells, transfectants of the HC1 1 cell

line expressing the human erbB-2 receptor, were cultured in

DMEM supplemented with 0.4 mg/mI neomycin analogue 0418

sulfate (geniticin; Life Technologies, Gaithersburg, MD; Ref.

34). flE or PA317 cells, transfected with recombinant retroviral

DNA, were cultured in DMEM with 0.6 mg/ml G418. Enriched

T cells were cultivated in DMEM with 10% FCS, 50 p.M

�3-mercaptoethanol (Life Technologies, Eggenstein, Germany),

10 m�t HEPES buffer (Life Technologies, Eggenstein, Germa-

ny), and 60 IU/ml human IL-2 (PBH, Hanover, Germany).

Animals. Male BDIX rats (about 200 g) and 6-8-week

old BALB/c female mice (H�2b) were obtained from Charles

River Breeding Laboratories (Sulzfeld, Germany).

3 A. Y. Nikitin, J. J. Jin, J. Papewalis, S. M. Prkopenko, A. Pozharisski,

and M. F. Rajewsky. Wild-type neu transgene counteracts mutant ho-mologue in malignant transformation of rat Schwann cells, submittedfor publication.

Construction of � Chain Fusion Genes. The retroviral

vector pLXSN has been described (33). The vector is named

based on the order of genetic elements that it encodes: L, long

terminal repeat promoter; X, cloning site for DNA insertion; 5,

5V40 promoter; and N, neomycin phosphotransferase gene

(neo). The Cl 1 antirat Neu antibody (35), the FRPS antihuman

erbB-2 antibody (36, 37), and the R-AK anti orthopox virus

antibody (a gift from C-P. Czerny, University of Munich,

Munich, Germany; Ref. 38) have been described. VH and VL

were derived from PCR amplifications using oligonucleotide

primers corresponding to the 5’ and 3’ consensus sequences of

the immunoglobulin V region and joined by a flexible linker, as

described (37). All scFvs were subcloned into pWWI52 as a

PstI-XbaI fragment (29). A fragment of the heregulin isoform

131 (70 amino acids) was cloned as described recently (39, 40).

The hinge/c fusion gene (41) was complemented by myc-tag,

which is recognized by the mAb Mycl-9E10 (42). All fusion

genes (Fig. I) were ligated as SalI-SrnaI fragments into the

retroviral vector pLXSN.

Enrichment of T Cells from Spleen Tissue. Spleencells from rats and mice were depleted of RBC by hypotonic

lysis with NH4CI and subsequently passed through a nylon-wool

syringe (43). The enriched cell preparation was analyzed by

FACS analysis and contained more than 85% T cells (TCR�),

less than 5% B cells (immunoglobulin�), and about 10% other

cells.

Retroviral Gene Transfer. Recombinant proviral DNA

was transfected into the fIE packaging cell line (32) by CaPO4

precipitation. The PA317 amphotropic packaging cell lines (33)

were infected by incubation with retroviral particles present in

the supernatants of transfected I1E cell cultures for 24 h in the

presence of 8 p.g/ml polybrene (Sigma Chemical Co., Deisen-

hofen, Germany). Virus titers produced by tzeo-resistant fIE and

PA3 17 cell clones were determined. NIH3T3 cells were infected

and neomycin-resistant cell clones were quantitated after addi-

tion of the neomycin analogue 041 8 at 1 mg/mI to the culture

medium. C196 cells (106) were infected with viral supernatants

of transfected f�E clones. They were grown for 24 h in medium

with IL-2 and polybrene. The virus-producing f�E cells had

been cultured for 24 h without G418 (Fig. 2). All supernatants

were filtered through a 0.22-p.m filter (Millipore, Eschborn,

Germany).

High-titer, neo-resistant CIE clones producing about I 0�

viral particles/ml in 24 h were selected and used for the cocul-

tivation experiments. Enriched primary T lymphocytes (106)

and l0� lIE cells (semiconfluent) were cocultured for 48 h in

complete medium (Fig. 2). Before the cocultivation, liE clones

were grown for 24 h in medium without neomycin. T cells were

separated from adherent I1E cells by careful pipetting and

grown in suspension with 5 p.g/ml PHA (Sigma), 60 IU/ml

recombinant human IL-2, and 8 p.g/ml polybrene. They were

used for the FACS analysis and cytotoxicity assays.

Antibodies and FACS Analysis. The following mAbs

were used for FACS analysis: mouse antirat CD4 (phyco-

erythrin labeled); mouse antirat CD8 (FITC labeled); mouse

antirat IgG (FITC labeled); goat antimouse TCR (FITC labeled);

and goat antimouse IgG (FITC labeled). All mAbs were pur-

chased from PharMingen (San Diego, CA). The mAb R73

specific for the rat TCR was a gift from T Hunig (University of



scFv(C1 1 ):myc:hinge:#{231}

B

jH�r�:eoHLTR�sv40 promoter

c Hs�� H LTR

4 scFv(R-AK):myc:hinge:#{231} -�

HRG:myc:hinge:#{231}

.. ., , ..r-�

retroviral DNA (+lnsert) animals

� surgery

cell lines (C196, Jurkat)

c:x:IXD

retroviralsupernatant

IL-2 � i�:i -

oootransduced cell lines

CaPO4

transfection

packaging celiline ciE

spleen

� homogenization

0000enriched T cells

activation:co-culture + PHA

+ IL-2

�a- transduced, activatedI lymphocytes

Clinical Cancer Research 1003

FACS-analysis (a myc)in vitro cytotoxicity

A

� LTR� VH I� VL � Hinge

Leader Linker myc-tag

�

� LTR HilVH �VL M�Hinge�

C

� LTR��L� VH � VL M� Hinge� � HSP� HLTR

D

� LTR��L�Here�uIin �ll M� Hinge � � H� neo HLTR

Fig. 1 Schematic representation of the fusion genes and the retroviral gene transfer vector pLXSN. scFv, fusion of cDNAs encoding the VH and VLto corresponding mAbs. VH and VL are joined by a short polypeptide linker sequence; L, leader sequence derived from an immunoglobulin

heavy-chain leader; Hinge, region of the CD8cx gene (29). CD3� is a signal-transducing component of the TCR-CD3 complex. Fusion gene productscould be detected by mAbs directed against the myc-tag epitope. A, scFv (Cl I) directed against the extracellular domain of the rat Neu protein. B,scFv (FRP5) directed against the extracellular domain of the human erbB-2 receptor. C, scFv (R-AK) directed against the Mr 14,000 surface proteinof the orthopox virus. D, human heregulin isoform [31 , the natural ligand to erbB-3 and erbB-4 receptors. The fusion genes are transcriptionallyregulated by the Moloney rnurine leukemia virus 5’ long terminal repeat (LTR). The vector also encodes a G418 resistance gene (neo), which isregulated by the SV4O promoter. Arrows, transcriptional start sites.

Fig. 2 Experimental strategyfor the transduction of T cells.Retroviral vectors harboring

the fusion gene inserts were in-

troduced in the packaging cell ________________line I1E by CaPO4 precipita-

tion. Enriched splenic primaryT lymphocytes from BDLX rats

or BALB/c mice were activatedwith PHA and IL-2 and trans-

duced by coculture with hightiter-producing QE clones. The _______T-cell line C196 was transducedby incubation with IL-2-sup-

plemented viral supernatants ofliE. The extent of transductionwas determined by FACS anal-

ysis. Cytotoxicity assays werecarried out.



I


Table I Determination of retroviral titers with NIH3T3 fibroblasts

Supernatants of flE PA317 clones were applied to NIH3T3 cells.G41 8-resistant NIH3T3 clones were selected and counted.

Retroviral constructs

Viral tite r (cfu/ml)

liE PA3 17

pLXSN 8 x i0� 3 x l0�

pLXSN + scFv(Cl l):rnyc:hinge:�pLXSN + scFv(FRP5):myc;hinge:�

7 x l0’�9 X l0�

3 X l0�2 x i0�

pLXSN + scFv(R-AT):myc:hinge:�

pLXSN + HGR:rnyc:hinge:�l0�

2 x l0�

4 X l0�

l0�

I �IA��Tir�FT:�D�T�� 100 101 102 1o� iO� 1�- � � � 104 100 101 102 103 io�

.�

Fluorescence iIltCIlSity

WUrzburg, Wurzburg, Germany). mAb Mycl-9E10 directed

against human myc was purified from supernatants of mouse

9E10 cells as described (42).

For FACS analysis, cells (5 X l0�-l0�) were reacted for 45

mm at 4#{176}Cwith S p.g Mycl-9ElO and antimouse IgG (FITC

labeled) or directly with a fluorescent-conjugated mAb, as de-

scribed by the manufacturers. Stained cells were washed twice

in PBS supplemented with 5% FCS and analyzed for antibody

binding by flow cytometry using a FACScan (Becton Dickin-

son, Heidelberg, Germany).

Cytotoxicity Assay. Cytotoxicity of T cells against

NV2Cd and HC1 lRl I I target cells were measured by 5tCr

release assays, as described recently (44). Effector T cells were

incubated at various E:T cell ratios with 7.5 X l0� 5tCr-labeled

target cells for 6 h at 37#{176}C.Thereafter, 100 p.1 cell-free super-

natants were removed and counted in a gamma counter. Specific

release was calculated using the formula:

% specific lysis =

(experimental release - spontaneous release). X100%

(maximal release - spontaneous release)

The maximal release is the release in 1 N HC1, and the sponta-

neous release is the release in the presence of medium only.

RESULTS

Construction of � Chain Fusion Genes. Three fusion

genes consisting of a tumor cell recognition function and the �

chain of the TCR complex were constructed (Fig. 1). The tumor

cell recognition functions are provided by the scFv Cl 1 directed

against the rat Neu protein and the scFv FRP5 directed against

human erbB-2. ScFv R-AK is directed against the Mr 14,000

surface protein of the orthopox virus and served as a control. All

scFvs are composed of the VH and VL of the mAb joined by a

1 5-amino acid linker sequence. One construct was generated in

which the sequence of human heregulin �1 (40), the natural

ligand of erbB-3 and erbB-4 receptors, served as a tumor cell

recognition domain. A myc-tag recognition epitope was in-

cluded into the constructs to facilitate the detection of their

expression with a myc-specific mAb. The myc-tag epitope is

recognized by the mAb 9ElO (42). This antibody was used to

determine the expression of the fusion proteins on the surface of

transduced T cells. A spacer domain, the hinge region, was

included in the constructs, because it was previously shown that

it enhances ligand recognition (29).

Fig. 3 Expression of � fusion genes on the cell surface of transduced

primary rat T lymphocytes. T cells were transduced by coculture with

lIE clones, producing the retroviral vector pLXSN (A) or recombinant

retroviral vectors harboring the fusion genes with scFv C 1 1 (B), scFv

FRP5 (C), scFv R-AK (D), and heregulin �3l (F). Cells were stained

with the mAb Mycl-9E10 directed against the myc-tag epitope of thefusion genes and with goat antimouse IgG (FITC labeled) and analyzed

by FACScan analysis (dark areas). Light areas, cells stained only with

the FITC-labeled antibody.

Retroviral Transduction of T Cells and Analysis of theExpression of the � Fusion Genes. The retroviral vector

pLXSN was used for gene transfer into T cells (Fig. 2). Enriched

naive lymphocytes from BDIX rats or BALB/c mice and the

established murine CTL line C196 were transduced with recom-

binant retroviruses. High-titer, virus-producing clones of the

packaging cell line fIE were selected by G4l8 treatment, and

the viral titers were determined (Table I ). Primary rat and

mouse T lymphocytes were transduced with recombinant or

parental retroviruses in the presence of the activating agents

PHA and IL-2 and coculture with retrovirus-producing I1E cells

(Fig. 2). After careful separation of the T cells from the adherent

liE cells, the expression of the fusion genes was determined by

FACS analysis. No further G418 selection was applied.

The mAb Mycl-9ElO was used to detect the expression of

the recombinant fusion proteins on the cell surface of the rat

lymphocytes (Fig. 3, B-E). Cells transduced with the retroviral

vector pLXSN do not express the epitope recognized by 9E10

and served as a negative control (Fig. 3A). The majority of the

activated T cells (>80%) expressed the gene constructs. The

mean fluorescence increased from about 18 obtained with con-

trol cells transduced with pLXSN to about 90-150 for the cells

transduced with � fusion genes. The transduced T-cell popula-

tions consisted of 60% ± 5% CD4� and 40% ± 5% CD8� T

cells (data not shown). Two different transduction strategies for

the scFv (Cli) fusion gene were compared: infection of T cells

with retroviral supernatants; and cocultivation with virus-pro-

ducing helper cells (Fig. 4). The transduction efficiency of naive

rat T lymphocytes, which were transduced by coculture with

virus-producing 1�1E cells (Fig. 4C), was 2-3-fold higher when

compared with T cells, which were treated with the retrovirus-

containing supernatants (Fig. 4B).

Primary mouse T lymphocytes were transduced with the

vector pLXSN or the retrovirus directing the expression of the



uninfected supernatant CO- culture

Fluorescence intensity

Fig. 4 Comparison of two different transduction strategies of naive rat T lymphocytes. T cells were treated with retroviral supernatant (B) or

cocultured with lIE clones producing the scFv C11:rnvc:hinge:� fusion gene (C). Untreated T cells served as controls (A). Cells were stained withmAb Mycl-9E10 directed against the myc-tag epitope of the fusion genes and with goat antimouse IgG (FITC labeled) and analyzed by FACScananalysis.

.0

EC

0)()

0)

�0

0)



I-0)E

0)U

0)

0)

scFv and heregulin fusion genes (Fig. 5). About 80% of the T

cells transduced with the fusion genes expressed the gene prod-

ucts on their cell surfaces (Fig. 5, B-E). The mean fluorescence

increased from about 6 obtained with cells transduced with

pLXSN (Fig. SA) to about 95 obtained with cells transduced

with the fusion genes.

The mouse CTL line Cl96 was infected by viral superna-

tants and selected by G4 I 8 treatment for 48 h (Fig. 1 ). There-

after, the expression of the fusion gene products on the cell

surface was determined by FACS analysis (Fig. 6). Transduc-

tion with the retroviral vector did not increase expression of the

myc-tag-containing proteins (Fig. 6A). As shown in Fig. 6, B-E,

infection of Cl96 cells with the recombinant retroviruses gen-

erated T-cell populations that expressed the fusion proteins to

nearly 100%. The mean fluorescence increased from about 3 for

cells transduced with pLXSN to about 30 for cells transduced

with recombinant vectors directing the expression of � fusion

proteins. Similar results were obtained when human Jurkat cells

were transduced by incubation with retroviruses containing su-

pernatants from PA317 cells (data not shown).

The expression of the scFv (Cli) fusion gene on primary

rat T cells was determined at different times after retroviral

infection. As shown in Fig. 7, the level of fusion gene expres-

sion did not change until day 10 (Fig. 7, D-F). Rat T cells

transduced with the retrovirus pLXSN served as a negative

control (Fig. 7, A-C).

Specific Cytotoxicity of Transduced Lymphocytes.

The cytotoxic activities of the transduced primary T lympho-

cytes from BDIX rats and the mouse T cell line Cl96 were

determined in vitro (Fig. 8). The Neu protein-expressing cancer

cell line NV2Cd derived from BDIX rats and the erbB-2-,

erbB-3-, and erbB-4-expressing mouse epithelial cell line

HCI 1RI I 1 were used as target cells. The release of 5tCr from

these cells was used as a measure of cell lysis. Syngeneic

primary T cells of BDIX origin transduced with the scFv (CII)

fusion gene lysed NV2Cd at E:T ratios of 4-100 (Fig. 8A). No

cell lysis could be measured when T cells expressing the scFV

(R-AK) fusion gene or the vector pLXSN were used. Similar

results were obtained with transduced Cl96 cells (Fig. 8C).

Fig. 5 FACS analysis of the expression of fusion genes on the cell

surface of transduced prinlary mouse T lymphocytes. T cells were

transduced by coculture with f�E clones producing the retroviral vector

pLXSN (A) or recombinant retroviral vectors harboring fusion genes

including scFv Cl I (B), scFv FRP5 (C), scFv R-AK (D), and heregulin

131 (E). Cells were stained with the mAb Mycl-9E10 directed against themyc-tag epitope of the fusion genes and with goat antimouse lgG (FITC

labeled) and analyzed by FACScan analysis (dark areas). Light areas,

cells stained only with the FITC-labeled antibody.

Naive rat lymphocytes transduced with scFv (FRP5) or heregu-

lin �3l fusion genes lysed HCI lRl 1 1 cells at E:T ratios of

0.8-100 with equal efficiencies (Fig. 8B).

Cytotoxic effects of Cl96 cells transduced with scFv

(FRP5) were determined with HCI IRI I 1 target cells at E:T

ratios of 100-20. For comparison, C196 cells transduced with

the heregulin �3l fusion gene lysed HC1 IR1 1 1 target cells at

E:T ratios of 5-100 (Fig. 8D).

The specific cytotoxic activity of the transduced primary T

cells and Cl96 cells shows that the strategies used here are able

to direct the cytotoxic activities toward tumor cells. The effi-

cient transduction of primary T cells and the grafting of target



pLXSN scFv (Cli)

0A

0.t � _� � -.-- 0� �

100 10� 102 i03 i04 100 1...

I ‘aI

102 10�i0�

0)

--0)U0)

0)

1k:100 101 102 103 1C

50

0

IA

(4

C

I�-��-‘---- od��

) 101 102 103 10’� 1� 1

F

J2 io� io�


�C�A(�J bk..:.] �::i� 100 101 102 io3 10” 100 10� 102 1o� i0� 100 101 iOZ iO) 104�80L��1 ____

100 101 102 io’ i0� i�#{176} 101 10� i03 10�


Fig. 6 Transduction of Cl96 and analysis of the expression of fusiongenes on the cell surface by FACScan analysis. C196 T cells wereincubated with supernatants of tiE clones producing the retroviralvector pLXSN (A) or recombinant retroviral vectors harboring fusiongenes with scFv Cli (B), scFv FRP5 (C), scFv R-AK (D), and heregulin�3l (E). Cells were stained with the mAb Mycl-9E10 directed against themyc-tag epitope of the fusion genes and with goat antimouse IgG (FITC

labeled) and analyzed by FACScan analysis (dark areas). Light areas,

cells stained only with the FITC-labeled antibody.

cell recognition specificity are important prerequisites for the

clinical use of CTLs in cancer treatment.

DISCUSSION

For the successful elimination of target cells, naturally

occurring T cells require the recognition of antigen peptides

presented in MHC complexes by the TCR and the interaction of

costimulatory molecules. The ability to alter the T-cell recog-

nition specificity using chimeric receptor genes may broaden the

prospects of adoptive cellular immunotherapy to include the

treatment of viral diseases and cancer. Several reports based on

this strategy have been published (16-19).

We are expanding on these observations and demonstrate

that it is also possible to manipulate primary T cells and to

transduce them with chimeric receptor genes. When these mol-

ecules are expressed on the surface of the T cells, they provide

them with the ability to lyse specific tumor cells in a MHC-

unrestricted manner. We used the natural ligand to the erbB-3

and erbB-4 receptors, heregulin �3l, and scFvs directed against

the rat Neu protein, against the erbB-2 receptor, and against a

viral surface protein as target structures to direct T-cell recog-

nition. These molecules were fused to the signal-transducing �

chain of the TCR complex; a CD8-a hinge region was intro-

duced to improve the ligand-binding property; and a myc

tag was introduced to follow the expression of the chimeric

molecules.

The fusion genes were transduced into T cells by retrovi-

ruses. The highest efficiency (80-100%) was obtained when

naive rat or mouse T cells were cocultured with retrovirus-

producing packaging cell lines in the presence of PHA and IL-2

as activators. The transduction efficiencies we obtained were

substantially higher than those described earlier. The transduc-

tion rate of CD4� and CD8� cells derived from peripheral

su

A01

100.

10� 1#{244}2

E

i03 1(


Fig. 7 FACS analysis of the expression of fusion genes on the cellsurface of transduced prirnary rat T lymphocytes at different times afterviral infection. T cells were transduced for 2 days by coculture with f1Eclones producing the retroviral vector pLXSN (A-C) or recombinantretroviral vectors harboring a fusion gene including scFv Cl 1 (D-F) andcultured without flE for 8 days. Cells were stained 2 days (A and D), 5days (B and E), and 10 days (C and F) after infection with the mAbMycl-9E10 directed against the myc-tag epitope of the fusion gene andwith goat antimouse IgG (FITC labeled) and analyzed by FACScan

analysis (dark areas). Light areas, cells stained only with the FITC-

labeled antibody.

blood lymphocytes was reported to be 6-8% (45, 46). Mavilio

et a!. (47) showed that after coculture with virus-producing

helper cells, 15% of human peripheral blood lymphocytes could

be transduced. Comparable transduction efficiencies were re-

cently reported for human thymic T-cell progenitor cells (65-

85%; Ref. 42). The transductions for the T-cell lines C196 and

Jurkat were performed by incubation with retroviral superna-

tants for 24 h. The efficiencies obtained (>80%) are comparable

with the ones reported by Staal et a!. (48).

We compared the transduction efficiency of naive T cells

after coculture and treatment with retroviral supernatants. The

transduction levels reached by coculture were 2-3-fold higher

when compared with the levels reached by transduction with

viral supematants. These differences were also observed with

Jurkat cells (48). The enhanced transduction efficiency obtained



NV2Cd HC11R1#11

A

>�

CC

C.)C)

C)I.)C.

60

50

40

30

20

I0

0

60

50

40

30

20.

I0

100 10 � 0.1

Err


��CD

100 ib i� 0’

Fig. 8 Specific MHC-unrestricted lysis of tumor cell lines by trans-

duced naive rat T cells (A and B) or T-cell line C196 (C and D). Primary

T lymphocytes were transduced with the retroviral fusion genes by

coculture with the retrovirus-producing packaging cell line lIE and thecell line C196 by incubation with retrovirus-containing supernatants. Rat

T cells or C196 cells were transduced with the scFv Cl I fusion gene(antirat ,ieu; #{149}),the scFv FRP5 fusion gene (antihuman erbB-2; A). thescFv R-AK fusion gene (antipox virus, control; A), the heregulin �3l

fusion gene (#{149}),or the retroviral vector pLXSN (0). Transduced T cellswere incubated with the 5tCr-labeled rat tumor cell line NV2Cd (A andB) or the 51Cr-labeled mouse erbB-2-, erbB-3-, and erbB-4-expressingtumor cell line HC1 IR1 1 1 (C and D) as targets in 6-h cytotoxicity

assays.

by cocultivation is most likely due to an optimal infection of

viruses as a result of cell-cell contact and to the high local virus

concentrations present between the cells. Furthermore, cell-cell

contacts could enhance the early events of T-cell activation by

costimulatory molecules, which are beneficial for retroviral

DNA integration.

The transduced genes include a myc-tag recognition

epitope, which allowed the direct detection of the fusion gene

products on the surface of infected cells. Because the myc-tag

epitope is part of the constant moiety of all fusion genes we

used, it is possible to analyze cells independently from their

recognition domains. The detection of recombinant scFv fusion

proteins was also shown with mAbs specifically directed against

the scFv ( 15, 1 7, 49). It is also possible to detect and to select

transduced cells by treatment with G418, but G4l8 might have

negative influences on the biological activity of naive T cells.

Also, the limited life span of naive T cells (50) makes G4l8

selection unfavorable. To avoid these disadvantages, we applied

the transduced T cells directly to in vitro cytotoxicity assays.

The results shown in Fig. 8 show that activated, transduced

naive T lymphocytes were able to lyse tumor cells expressing

specific target structures. Transduced T cells properly process

the � fusion molecules, and cytolysis is induced on engagement

of the extracellular tumor cell recognition domains. These re-

sults complement recent studies in which chimeras of the ‘y or #{128}

chains have been joined to extracellular CD4, CD8, IL-2 recep-

tor, and CD 16 domains and expressed in transduced T lympho-

cytes and basophiles (18, 51, 52).

In this report, an efficient procedure is described to gener-

ate genetically manipulated naive T cells with a grafted recog-

nition specificity. Our results show that the cytolytic effector

machinery can be redirected toward cells expressing the Neu/

erbB-2, erbB-3, and erbB-4 receptors. These receptors are im-

portant in breast, ovarian, gastric, and colon cancer. The effi-

cient provision of tumor cell-specific recognition specificity to

primary T cells moves the strategy closer to become a useful

therapeutic tool for cancer treatment.

REFERENCES

1 . Greenberg, P. D. Adoptive T cell therapy of tumors: mechanismsoperative in the recognition and elimination of tumor cells. Adv. Im-

munol., 49: 281-355, 1991.

2. Old, L. J. Cancer immunology: the search for specificity. Cancer

Res., 41: 361-375, 1981.

3. Van der Bruggen, P., Traversari, C., Chomez, P., Lurquin, C., De

Plaen, E.. Van den Eynde, B., Knuth, A., and Boon, T. A gene encodingan antigen recognition by cytolytic T lymphocytes an a human mela-

noma. Science (Washington DC), 254: 1643-1647, 1991.

4. W#{246}lfel, T., Hauer, M.. Scheider, J., Serrano, M., Wdlfel, C.,

Klehmann-Hieb, E., De Plaen, E., Hankeln, T. Meyer zum BUschen-felde, K-H., and Beach, D. A p16�’”’-insensitive CDK4 mutant tar-geted by cytolytic T lymphocytes in a human melanoma. Science

(Washington DC), 269: 1281-1284, 1995.

5. Rosenberg, S. A. Immunotherapy and gene therapy of cancer. Cancer

Res. 51 (Suppl.): 5074s-5079s, 1991.

6. Chang, A. E., and Shu, S. Immunotherapy with sensitized lympho-

cytes. Cancer Invest., 10: 357-369, 1992.

7. Kast, M. W., Offringa, R., Peters, P. J., Voordouw, A. C., Meloen,

R. H., van der Eb, A. J., and Melief, C. J. M. Eradication of adenovirus

El-induced tumors by EIA-specific cytotoxic T lymphocytes. Cell, 59:

603-614, 1989.

8. Aebersold, P., Hyatt, C., Johnson, S., Hines, K., Korcak, L., Sanders,

M., Lotze, M., Topalian, S., Yang, J., and Rosenberg, S. A. Lysis of

autologous melanoma cells by tumor infiltrating lymphocytes: associa-

tion with clinical response. J. Nail. Cancer Inst., 83: 932-937, 1991.

9. Fox, B. A., and Rosenberg, S. A. Heterogeneous lymphokine acti-

vated killer cell precursor populations. Development of a monoclonal

antibody that separates two populations of precursors with distinct

culture requirements and separate target-recognition repertoires. Cancer

Immunol. Immunother., 29: 155-166, 1989.

10. Rosenberg, S. A., Aebersold, P., Cornetta, K., Kasid, A., Morgan,

R. A., Moen, R., Karson, E. M., Lotze, M. T., Yang, J. C., Topalian,S. L., Merino, M. J., Culver, K., Miller, A. D., Blaese, R. M., and

Anderson, W. F. Gene transfer into humans-immunotherapy of pa-tients with advanced melanoma, using tumor-infiltrating lymphocytesmodified by retroviral gene transduction. N. EngI. J. Med.. 323.’ 601-603, 1990.

I I . Davis, M. M., and Bjorkman, P. J. T-ceIl antigen receptor genes and

T-cell recognition. Nature (Land.), 334: 395-402, 1988.

12. Klausner, R. D., and Samelson, L. E. T cell antigen receptoractivation pathways: the tyrosine kinase connection. Cell, 64: 875-878,

1991.

13. Frank, S. J., Niklinska, B. B., Orloff. D. G., Mercep, M., Ashwell,

J. D.. and Klausner, R. D. Structural mutations of the T cell receptor �

chain and its role in T cell activation. Science (Washington DC), 249:

174-177, 1990.

14. Weiss, A. T cell antigen receptor signal transduction: a tale of tails

and cytoplasmic protein tyrosine kinases. Cell, 73: 209-213, 1993.

15. Weiss, A., and Littrnan, D. R. Signal transduction by lymphocyte

antigen receptors. Cell, 76: 263-274, 1994.




16. Brocker, T., and Karjalainen, K. Signals through T cell receptor-cchain alone are insufficient to prime resting T lymphocytes. J. Exp.

Med., 181: 1653-1659, 1995.

17. Moritz, D., Wels, W., Mattern, J., and Groner, B. Cytotoxic T

lymphocytes with a grafted recognition specificity for ERBB2-express-

ing tumor cells. Proc. Nail. Acad. Sci. USA, 91: 4318-4322, 1991.

18. Romeo, C., and Seed, B. Cellular immunity to HIV activated byCD4 fused to T cell or Fe receptor polypeptides. Cell, 64: 1037-1046,

1991.

19. Eshhar, Z., Waks, T., Gross, G., and Schindler, D. G. Specificactivation and targeting of cytotoxic lymphocytes through chimericsingle chains consisting of antibody-binding domains and the -�‘ or �

subunits of the immunoglobulin and T-cell receptors. Proc. Nail Acad.Sci. USA, 90: 720-724, 1993.

20. Slamon, D. J., Godolphin, W., Jones, L. A., Holt, J. A., Wong,

S. G., Keith, D. E., Levin, W. J., Stuart, S. G., Udove, J., and Ullrich,

A. Studies of the Her-2lneu proto-oncogene in human breast and ovar-ian cancer. Science (Washington DC), 244: 707-712, 1989.

21. Semba, K., Kamata, N., Toyoshima, K., and Yamamoto, T. Av-erbB-related protooncogene, c-erbB-2, is distinct from the c-erbB-1/

epidermal growth factor-receptor gene and is amplified in a humansalivary gland adenocarcinoma. Proc. Natl. Acad. Sci. USA, 82: 6497-6501, 1985.

22. Fukushige, S. I., Matsubara, K., Yoshida, M., Sasaki, T., Semba, K.,

Toyoshima, K., and Yamamoto, T. Localization of a novel v-erbB-

related gene, c-erbB-2, on human chromosome 17 and its amplificationin a gastric cancer cell line. Mol. Cell. Biol., 6: 955-958, 1986.

23. Peles, E., and Yarden, Y. Neu and its ligands: from an oncogene to

neural factors. BioEssays, 15: 815-824, 1993.

24. Hynes, N. E., and Stern, D. F. The biology of erbB-2/neu/Her-2 and

its role in cancer. Biochim. Biophys. Acta, 1198: 165-184, 1994.

25. Kraus, M. H., Issing, W., Mild, T., Popescu, N. C., and Aaronson,S. A. Isolation and characterization of ERBB3, a third member of the

ERBB/epidermal growth factor receptor family: evidence for overex-pression in a subset of human mammary tumors. Proc. Nail. Acad. Sci.

USA, 86: 9193-9197, 1989.

26. Lernoine, N. R., Barnes, D. M., Hollywood, D. P., Hughes, C. M.,

Smith, P., Dublin, E., Prigent, S. A., Gullick, W. J., and Hurst, H. C.Expression of the ERBB3 gene product in breast cancer. Br. J. Cancer,

66: 1116-1121, 1992.

27. Plowman, G. D., Culouscu, J. M., Whitney, G. S., Green, J. M.,

Canton, G. M., Foy, L., Neubauer, M. G., and Shoyab, M. Ligand-

specific activation of Her4/pl8OerbB4, a fourth member of the epider-

mal growth factor receptor family. Proc. NatI. Acad. Sci. USA, 90:

1746-1750, 1993.

28. Plowman, G. D., Green, J. M., Culouscu, J. M., Carlton, G. M.,Rothwell, V. M., and Buckley, S. Heregulin induces tyrosine phospho-rylation of HER4/pl8OerbB4. Nature (Lond.), 366: 473-475, 1993.

29. Moritz, D., and Groner, B. A spacer region between the single chain

antibody- and the CD3 c-chain domain of chimeric T-cell receptor

components is required for efficient ligand binding signaling activity.Gene Ther., 2: 539-546, 1995.

30. Eichmann, K., Ehrfeld, A., Falk, I., Goebel, H., Kupsch, J.,Reimann, A., Zgaga-Gnesz, A., Saizawa, K. M., Yachelini, P., andTomonari, K. Affinity enhancement and transmembrane signalling areassociated with distinct epitopes on the CD8 cx�3 heterodimer. J. Irnmu-nol., 147: 2075-2081, 1991.

3 1 . Karasuyama, H., and Melchers, F. Establishment of mouse cell linewhich constitutively secrete large quantities of interleucin 2, 3, 4, or 5,

using modified eDNA expression vectors. Eur. J. Irnmunol., 18: 97-104,1988.

32. Morgenstern, J. P., and Land, H. Advanced mammalian gene trans-

fer: high titer retroviral vectors with multiple drug selection markers anda complementary helper-free packaging cell line. Nucleic Acids Res.,

18: 3587-3596, 1990.

33. Miller, A. D., and Rosman, G. J. Improved retroviral vectors for

gene transfer and expression. Biotechniques, 7: 980-990, 1989.

34. Hynes, N. E., Taverna, D., Harwerth, I. M., Ciardiello, F., Salomon,

D. S., Yamamoto, T., and Groner, B. Epidermal growth factor, but not

c-erbB-2, activation prevents lactogenic hormone induction of the �3-ca-

sein gene in mouse mammary epithelial cells. Mol. Cell. Biol., 10:

4027-4034, 1990.

35. Yarden, Y. Agonistic antibodies stimulate the kinase encoded bythe neu protooncogene in living cells but the oncogenic mutant isconstitutively active. Proc. Natl. Acad. Sci. USA, 87: 2569-2573, 1990.

36. Harwerth, I. M., Wels, W., Marie, B. M., and Hynes, N. E. Mono-clonal antibodies against the extracellular domain of the erbB-2 receptor

function as partial ligand agonist. J. Biol. Chem., 267: 15160-15167,

1992.

37. Wels, W., Harwerth, I. M., Mueller, M., Groner, B., and Hynes,

N. E. Selective inhibition of tumor cell growth by a recombinant

single-chain antibody-toxin specific for the erbB-2 receptor. CancerRes., 52: 6310-6317, 1992.

38. Meyer, H., Osterrieder, N., and Czerny, C-P. Identification of

binding sites for neutralizing monoclonal antibodies on the 14 kDa

fusion protein of orthopox viruses. Virology, 22: 778-783, 1994.

39. Holmes, W. E., Sliwkowski, M. X., Akita, R. W., Henzel, W. J.,Lee, J., Park, J. W., Yansura, D., Abadi, N., Raab, H., and Lewis, G. D.Identification of heregulin, a specific activator of pl85erbB2. Science

(Washington DC), 256: 1205-1210, 1992.

40. Jeschke, M., Wels, W., Dengler, W., Irnber, R., Stocklin, E., and

Groner, B. Target inhibition of tumor-cell growth by recombinant

heregulin-toxin fusion proteins. Int. J. Cancer, 60: 730-739, 1995.

41. Weissmann, A. M., Baniyash, M., Hou, H., Samelson, L. E.,

Burgess, W. H., and Klausner, R. D. Molecular cloning of the � chain of

the T cell antigen receptor. Science (Washington DC), 239: 1018-1021,

1988.

42. Evan, G. I., Lewis, G. K., Ramsay, G., and Bishop, J. M. Isolation

of monoclonal antibodies specific for human c-myc proto-oncogene

product. Mol. Cell. Biol., 5: 3610-3616, 1985.

43. Coligan, J. E., Kruisbeek, A. M., Margulies, D. H., Shevach, E. M.,

and Strober, W. (eds.). Current Protocols in Immunology, Vol. 2, pp.

3.2.1-3.2.3. New York: J. Wiley & Sons, Inc., 1993.

44. Brett, S. J., Mazurov, A. V., Charles, I. G., and Tite, J. P. Theinvasin protein of Yersinia spp. provides co-stimulatory activity tohuman T cells through interaction with �3l integrins. Eur. J. Imrnunol.,23: 1608-1614, 1993.

45. Morecki, S., Karson, E., Cornetta, K., Kasid, A., Aebersold, P.,Blaese, M., Anderson, W. F., and Rosenberg, S. A. Retrovirus-mediatedgene transfer into CD4� and CD8� human T cell subsets derived from

ilL and PBMC. Cancer Immunol. Immunother., 32: 342-352, 1991.

46. Culver, K., Cornetta, S., Rorgan, S., Morecki, S.. Aebersold, P.,Blaese, M., Anderson, W. F., and Rosenberg, S. A. Lymphocytes as

cellular vehicles for gene therapy in mouse and man. Proc. Nat!. Acad.

Sci. USA, 88: 3155-3159, 1991.

47. Mavilio, F., Ferrari, G., Rossini, S., Nobili, N., Bonini, C., Casorata,

G., Traversari, C., and Bordigon, C. Peripheral blood lymphocytes astarget cells of retroviral vector-mediated gene transfer. Blood, 50:

1988-1997, 1994.

48. Stan!, F. J. T., Rees, P. C. M., Weijer, K., and Spits, H. Develop-

rnent of retroviral cells into mature thymocytes. Int. Immunol., 7:

1301-1309, 1995.

49. Hwu, P., Yang, J. C., Cowherd, R., Treisman, J., Shafer, G. E.,Eshhar, Z., and Rosenberg, S. A. In vitro antitumor activity of T cells

redirected with chimeric antibody/T cell receptor genes. Cancer Res.,

55: 3369-3373, 1995.

50. Sprent, J., and Tough, D. F. Lymphocyte life-span and memory.

Science (Washington DC), 265: 1395-1 397, 1994.

5 1 . Letourneur, F., and Klausner, R. D. T-cell and basophil activation

through the cytoplasmic tail of T-celI-receptor � family proteins. Proc.

NatI. Acad. Sci. USA, 88: 8905-8909, 1991.

52. Irving, B. A., and Weiss, A. The cytoplasmic domain of the T cell

receptor � chain is sufficient to couple to receptor-associated signal

transduction pathways. Cell, 64: 891-901, 1991.



1996;2:1001-1008. Clin Cancer Res U Altenschmidt, R Kahl, D Moritz, et al. erbB-4 receptors by genetically targeted naive T lymphocytes.Cytolysis of tumor cells expressing the Neu/erbB-2, erbB-3, and

Updated version

http://clincancerres.aacrjournals.org/content/2/6/1001

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

[email protected] 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://clincancerres.aacrjournals.org/content/2/6/1001To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Cytolysis of Tumor Cells Expressing the Neu/erbB-2, erbB-3 ...Vo!. 2, 1001-1008, June 1996 Clinical Cancer Research 1001 Cytolysis of Tumor Cells Expressing the Neu/erbB-2, erbB-3,