Proc. Natl. Acad. Sci. USA
`Vol. 86, pp. 10024-10028, December 1989
`Immunology
`
`Expression of immunoglobulin-T-cell receptor chimeric molecules
`as functional receptors with antibody-type specificity
`(chimeric genes/antibody variable region)
`GIDEON GROSS, TOVA WAKS, AND ZELIG ESHHAR*
`Department of Chemical Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel
`
`Communicated by Michael Sela, July 13, 1989 (received for review June 18, 1989)
`
`ABSTRACT
`To design and direct at will the specificity of
`T cells in a non-major histocompatibility complex (MHC)-
`restricted manner, we have generated and expressed chimeric
`T-cell receptor (TcR) genes composed of the TcR constant (C)
`domains fused to the antibody's variable (V) domains. Genomic
`expression vectors have been constructed containing the re-
`arranged gene segments coding for the V region domains of the
`heavy (VH) and light (VL) chains of an anti-2,4,6-trinitrophenyl
`(TNP) antibody (SP6) spliced to either one of the C-region gene
`segments of the a or (3 TcR chains. Following transfection into
`a cytotoxic T-cell hybridoma, expression of a functional TcR
`was detected. The chimeric TcR exhibited the idiotope of the
`Sp6 anti-TNP antibody and endowed the T cells with a non-
`MHC-restricted response to the hapten TNP. The transfectants
`specifically killed and produced interleukin 2 in response to
`TNP-bearing target cells across strain and species barriers.
`Moreover, such transfectants responded to immobilized TNP-
`protein conjugates, bypassing the need for cellular processing
`and presentation. In the particular system employed, both the
`TNP-binding site and the Sp6 idiotope reside almost exclusively
`in the VH chain region. Hence, introduction into T cells of TcR
`genes containing only the VHSp6 fused to either the Ca or C(3
`was sufficient for the expression of a functional surface recep-
`tor. Apparently, the VHCa or VHCP chimeric chains can pair
`with the endogenous ( or a chains ofthe recipient T cell to form
`a functional a(3 heterodimeric receptor. Thus, this chimeric
`receptor provides the T cell with an antibody-like specificity
`and is able to effectively transmit the signal for T-cell activation
`and execution of its effector function.
`
`Antigen recognition by T-cell receptors (TcRs) differs from
`that of antibodies in that antibodies interact with antigens in
`their native state with relatively high affinity, whereas TcRs
`recognize fragmented antigens bound to cell surface class I
`and class II major histocompatibility complex (MHC) mole-
`cules (1). The differences in specificity of TcRs versus
`antibodies are not, however, evident from differences in their
`molecular structure. In fact, both molecules consist of two
`disulfide-linked polypeptide chains referred to as a and ,3 for
`the TcR and heavy (H) and light (L) for antibodies, each
`containing constant (C) and variable (V) domains. The anti-
`gen-binding site of both antibodies and TcR is encoded by a
`V-region exon, formed by rearrangment of V, diversity and
`joining gene segments (reviewed in ref. 2). In the case of
`immunoglobulins, x-ray data have shown that the V regions
`are loosely connected to the C regions (3), and, because of
`their high degree of homology with immunoglobulin, it is
`highly likely that TcRs are constructed similarly. Therefore,
`from a structural standpoint, it should be possible to confer
`antibody specificity on T cells by removing TcR V regions
`and replacing them with antibody V regions. Such a "chi-
`
`The publication costs of this article were defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertisement"
`in accordance with 18 U.S.C. §1734 solely to indicate this fact.
`
`meric TcR" (cTcR) would contain the extracellular C region,
`the transmembrane segment, and the cytoplasmic domains of
`normal TcRs and should therefore be able to function nor-
`mally to induce T-cell proliferation, interleukin production,
`and target cell lysis.
`Spontaneous transcription of an aberrantly joined IgVH
`gene and a TcR JaCa gene resulting from site-specific
`chromosome 14 inversion in human T-cell tumors was re-
`ported (4-6); however, no protein product was detected.
`Chimeric fusion proteins have also been produced in myelo-
`mas by the introduction of the TcR C exons between the VK
`and CK exons (7). More recent reports have shown that a
`chimeric protein containing the TcR V,, domain and the
`immunoglobulin C domain can be synthesized in myeloma
`cells. This protein associates with normal L chains to form a
`secreted tetramer (8). Attempts to assemble and secrete
`similar chimeric protein containing the VtC,-, and VC3CK have
`not been successful (9).
`The studies described above all reported the construction
`of nonfunctional chimeric Ig-TcR proteins. In this paper we
`describe the construction and functional expression in T cells
`of chimeric TcR genes made by recombining the immuno-
`globulin VH and VL rearranged gene segments to the C-region
`exons of the TcR a and 13 chains. The resulting cTcR is
`expressed on the surface of cytotoxic T lymphocytes, rec-
`ognizes antigen in a non-MHC-restricted manner, and effec-
`tively transmits the transmembrane signal for T-cell activa-
`tion.
`
`MATERIALS AND METHODS
`Construction of Chimeric Genes: Isolation of VL and VH
`Gene Segments. Genomic clones containing the rearranged
`VJK (TK1 plasmid) and VDJH (pR-Sp6 plasmid) derived from
`the IgM (AK) anti-2,4,6-trinitrophenyl (TNP)-producing Sp6
`hybridoma (10-11) were kindly provided by G. Kohler (Max-
`Planck-Institut fur Immunbiologie). The 2.2-kilobase-(kb)
`Pst I fragment containing L-VJK from TK1 and the 1.7-kb Xba
`I fragment containing L-VDJH were isolated, and each was
`subcloned into pUC19.
`Isolation of TcR Ca and C.6 Clones. Two genomic clones,
`ACa2 and AC,611, which contained all of the C exons and a
`large portion of the upstream intron, were isolated from a
`mouse embryonic library in Charon 4A (12) using Ca and C/3
`probes derived from the TT11 and 86T1 cDNA clones,
`respectively, kindly provided by M. M. Davis (Stanford
`University School of Medicine). A 9.0-kb BamHI fragment
`containing all Ca exons plus 5' 1.0 kb and a 6.3-kb BamHI
`
`Abbreviations: TcR, T-cell receptor; cTcR, chimeric TcR; MHC,
`major histocompatibility complex; V, variable; C, constant; H,
`heavy; L, light; mAb, monoclonal antibody; TNP, 2,4,6-trinitro-
`phenyl; IL-2, interleukin 2; APC, antigen-presenting cell; LTR, long
`terminal repeat; CTL, cytotoxic T lymphocyte; RSV, Rous sarcoma
`virus.
`*To whom reprint requests should be addressed.
`
`10024
`
`Downloaded from https://www.pnas.org by 136.226.53.3 on February 5, 2023 from IP address 136.226.53.3.
`
`UPenn Ex. 2060
`Miltenyi v. UPenn
`IPR2022-00855
`
`

`

`10025
`
`Immunology: Gross et al.
`Grossetal.~~~~Proc.
`Immunology:
`Natl. Acad. Sci. USA 86 (1989)
`fragment containing all of the C,32 exons plus 5' 1.9 kb were
`each subcloned into the BamH1 site of pUC19.
`Cloning of the Chimeric Genes into Expression Vectors. The
`vectors used for construction and transfection of the chimeric
`Ig-TcR genes were pRSV2neo and pRSV3gpt (kindly pro-
`vided by DNAX). These are derivatives of pSV2neo (13) and
`pSV3gpt (14) vectors into which the long terminal repeat
`(LTR) of the Rous sarcoma virus (RSV) was introduced. To
`construct the VL-containing chimeric vectors, we isolated
`from the TKd plasmid a 0.9-kb BgI II-BamHI fragment by
`partial Bgl II digestion followed by BamHI digestion. This
`fragment, containing the complete L-VJK, segment, was in-
`serted into the BamHI site of pRSV3gpt to produce
`pRSV3VL, in which L-VJK, is in the correct transcription
`orientation with regard to the RSV-derived LTR. Into the 3'
`BamHI of the insert, the 9.0-kb BamHI containing Ca or the
`6.3-kb BamHI fragment containing C/32 was introduced to
`generate pRSV3VLCa or pRSV3VLCI3, respectively (Fig. 1).
`The VH-containing chimeric vectors were constructed by
`introducing in the correct orientation into the BamHI site of
`pRSV2neo a 1.6-kb BcI I-BarnHI fragment containing the
`L-VDJH of Sp6. The remaining 3' BamHI site was then used
`to insert the BamHI Ca and Cf3 fragments described above to
`CQIRCIIICaM
`B
`Bc LVDJH
`B
`Cal
`I r
`Ii
`
`n F-i
`
`produce pRSV2VHCa and pRSV2VHC/3 (Fig. 1), respec-
`tively. For both VL and VH constructions, we used naturally
`occurring restriction sites complementary to the BamHI site
`(BgI II in VL and Bcl I in VH), which reside 25 and 45 base
`pairs, respectively, upstream from the starting ATG codon of
`both leader peptide exons. No other ATG is present between
`the LTR promoter and the starting codons in either case. This
`ensures that the original starting methionine will be con-
`served and that no transcriptional control elements of the
`immunoglobulin genes will be included.
`Transfection. MD.45, an H-2Db allospecific cytotoxic T-
`cell hybridoma (15), was transfected by protoplast fusion as
`described by Ochi et al.
`(16). Transfectants receiving
`pRSV2neo- or pRSV3gpt-based vectors were selected for
`growth in the presence of G418 (2 mg/ml) (GIBCO) or
`mycophenolic acid (1 gg/ml) (GIBCO), hypoxanthine (15
`~tg/ml) (Sigma), and xanthine (200 ug/ml) (Sigma), respec-
`tively.
`RNA Analysis. Cytoplasmic RNA was prepared from trans-
`fected cells (17). Northern blots were performed by standard
`techniques and probed with the following probes labeled by
`nick-translation: Ca, 0.55-kb Nco I fragment of TT11 plas-
`mid; Cf3, 0.3-kb Xho I-Nco I fragment of 86T1 plasmid; VH,
`1.7-kb Xba I fragment of pR-Sp6 (11); VL, 0.9-kb Bgl II
`fragment of TK1 (10).
`Imunoblotting. Pellets containing 2 x i01 cells were lysed
`in 200 gl of solution containing 1% Nonidet P-40, 10 mM
`EDTA, 1 mM phenylmethylsulfonyl fluoride (Sigma), 10 tgg
`of aprotinin per ml (Sigma), and 10 units of leupeptin per ml
`(Sigma) in 10 mM Tris, pH 8.0/0. 15 M NaCl. After 30 min at
`00C and centrifugation at 12,000 x g, 50,p1 of each superna-
`tant was boiled for 2 min in sample buffer containing 10 mM
`iodoacetic acid. NaDodSO4/PAGE through 10% gels was
`performed by the method of Laemmli (18). Separated pro-
`teins were blotted onto nitrocellulose filters (19) and allowed
`to react with the anti-Sp6 idiotypic monoclonal antibody
`(mAb) 20.5 (20) (provided by G. Koher), followed by affin-
`ity-purified '251I-labeled goat anti-mouse Fab' antibody.
`Interleukin 2 (IL-2) Secretion Assays. Transfectants were
`incubated with various target cells and antigens in Dulbecco's
`modified Eagle's medium containing 10% fetal calf serum for
`18-36 hr. IL-2 secretion was assayed by using an IL-
`2-dependent cell line and the methyltetrazolium staining (21).
`One unit of IL-2 is defined as an inverse of the dilution that
`supported 50% growth of the cytotoxic T lymphocyte CTL-L
`indicator cells.
`Cytotoxicity Assay. The 51Cr-release assay (15) was used to
`evaluate the ability of the transfectants to lyse specific target
`cells. Target cells were modified by TNP using 10 mM
`2,4,6-trinitrobenzenesulfonic acid as described (22). Killing
`assays were performed at different ratios of effector to
`5tCr-labeled target cells for 4-8 hr.
`
`RESULTS
`Construction and Expression of Chimeric TcR Genes. To
`produce a chimeric TcR with an antigen-binding site of a
`given antibody molecule, we have combined each of the
`rearranged L-VDJH and L-VJK, gene segments of the anti-
`TNP mAb Sp6 with gene segments encoding the C region of
`either one of the a or p TcR chains. In the final expression
`vector, transcription is driven by the RSV LTR inserted 5' to
`the immunoglobulin leader exon, next to the starting ATG
`codons. Fig. 1 depicts schematically the different chimeric
`genes and the expression vectors used. Because we could not
`predict whether a certain combination Of VH or VL combined
`with either Ca or C.8 will result in better pairing with the
`complementary chain, we constructed all four combinations
`of chimeric geneS-VHCa, VHCI3, VLCa, and VLCP.- Plas-
`mids were constructed so that each cTcR chain was associ-
`
`pBR 322 ori
`
`SV40 ori
`
`Bc LVDJH
`
`BR
`
`R
`
`WI c~Iq/3g I3
`
`B
`
`pBR322 ori
`
`SV40 ori
`
`1kb
`
`BgLVJK8
`
`Cai
`
`CaR aaIB
`
`a-
`
`pBR 322 cri
`
`SV40 or1
`
`Bg LVJK BR
`I
`Ya-
`
`R
`
`I
`
`ciclo.9f
`
`fI
`
`B
`
`'q
`
`pR
`
`3V
`
`p BR32 2 or'
`
`SV40Oor,
`
`1k b
`
`FIG. 1.
`
`Scheme of the chimeric antibody/TcR genomic con-
`structs inserted into the BamHI (B) site of the pRSV expression
`vectors. The boxes along the linear insert represent (from left to
`right) the exons coding for the leader (L), rearranged V region of the
`H chain (VDJH) or KL chain (VJK,), and the four C domains of the TcR
`chains. SV4O, simian virus 40.
`a and
`
`Downloaded from https://www.pnas.org by 136.226.53.3 on February 5, 2023 from IP address 136.226.53.3.
`
`UPenn Ex. 2060
`Miltenyi v. UPenn
`IPR2022-00855
`
`

`

`10026
`
`Immunology: Gross et al.
`
`Proc. Natl. Acad. Sci. USA 86 (1989)
`
`Fig. 2) hybridized only with VL probes and the transfectants
`that received the VHCa (c.2) or VHCI3 (d.9) hybridized only
`with the VH Sp6 probe. Almost all of the transfectants that
`received both chimeric genes (represented by clones e, f, g,
`and h) transcribed both RNAs into transcripts of the pre-
`dicted size (1.4 kb for VLCB or VHCP and 1.7 kb for VLCa
`or VHCa). The level of expression of chimeric RNA was far
`greater than the expression of the endogenous TcR a and f8
`chains of the same cells (data not shown).
`Expression of the chimeric polypeptide chains was ana-
`lyzed in cell lysates by immunoblotting using the anti-Sp6
`idiotypic mAb 20.5 (Fig. 2B). The idiotope recognized by this
`antibody is sensitive to reduction (G.G. and Z.E., unpub-
`lished data) and is expressed exclusively by the VH of the Sp6
`,u chain: the antiidiotypic mAb reacted only with transfec-
`tants that received chimeric genes containing VH either alone
`(such as c.2 and d.9, shown after longer exposure of the blot,
`right panel of Fig. 2B) or together with VL chimeric genes
`(e-h, Fig. 2B). The doubly transfected cells expressed a
`major broad band (apparently composed of two major bands
`of 85 and 90 kDa) that corresponds to heterodimeric proteins,
`bands of higher molecular mass of about 160 kDa, and higher
`molecular mass bands that most likely correspond to het-
`erotetrameric and higher complexes of the chimeric receptor.
`Bands of about 38 kDa (in recipients of the VHCa chimeric
`genes) and 42 kDa (in VHCI3 recipients) are also apparent,
`representing the non-S-S linked, monomer chimeric chains
`present in the cell lysates. Similar blots of gels electropho-
`resed under reducing conditions or developed with a radio-
`iodinated anti-mouse Fab' antibody failed to give any signal.
`The heterogeneity in size of the chimeric a/,8 heterodimer
`that was also observed in immunoprecipitation analysis of
`surface labeled receptors, (ref 24 and unpublished data) is
`most likely due to the formation of heterodimers that are
`formed by pairing between the chimeric chains and the
`complementary endogenous a or 18 chains of the MD.45
`hybridoma. The intensity of the different bands reflects the
`amount of the chimeric receptor in the cell lysates and the
`degree of reactivity with the antiidiotypic antibody.
`Functional Expression of the Chimeric TcR Genes. To
`examine whether the chimeric TcR genes code for a func-
`tional receptor, we tested the ability of the various transfec-
`tants to produce lymphokines in response to TNP-modified
`cells and to kill target cells bearing TNP groups on their
`surface. When TNP-A.20, (a TNP-modified B-cell lymphoma
`from BALB/c mice), was used to stimulate the various
`transfectomas, almost all of the double transfectants pro-
`duced IL-2 (Fig. 3). No production of IL-2 was evident when
`unmodified A.20 cells were used as stimulators (data not
`
`CELL TRANSFECTED
`1st
`LINE
`
`CONSTRUCT
`2nd
`
`MD.45
`
`GTA .a
`GTA .b
`
`GTA .c
`
`GTA .d
`
`GTA.e
`
`GTA .f
`
`GTA. g
`
`GTA.h
`
`-
`
`gpt-VLCa
`gpt -VLCO
`
`neo - VHCa
`
`neo -VHCO
`gpt -VLCa
`
`gpt - VLCO
`
`neo -VHCa
`
`neo-V"Co
`
`-
`
`-
`-
`
`-
`
`-
`neo-VHCO
`neo -VH Ca
`
`gpt -VL CR
`
`gpt -VL Ca
`
`ated with a different selectable marker (gpt or neor genes)
`than was its complementary chain, to allow selection of
`transfectants coexpressing both chimeric genes. To express
`these chimeric genes in T cells, we transfected them into the
`MD.45 murine hybridoma. The MD.45 hybridoma specifi-
`cally lyses H-2Db target cells (15) and, upon stimulation with
`either T-cell mitogens or appropriate target cells, produces
`IL-2, interleukin 3, and granulocyte/macrophage colony-
`stimulating factor. The hybridoma subclone used in this study
`expresses both of the a and ,B TcR chains of its CTL parent
`and only the 13 chain of BW5147 (23). We first transfected into
`MD.45 each of the four chimeric constructs independently
`and analyzed the resulting transfectants for expression of
`chimeric transcripts. We then retransfected the positive
`clones with the complementary constructs.
`In the first round of transfections with single chimeric
`genes, out of 48 wells seeded for each transfection, growth
`was seen in 21 of those that received VLCa (termed GTA.a),
`15 that received VLCB (termed GTA.b), 9 that received
`VHCa (termed GTA.c), and 13 that received VHCP (termed
`GTA.d). Clones expressing high RNA levels of each series
`were then transfected with the complementary construct. In
`all, out of 3 x 107 cells transfected, 54 independent trans-
`fectants were obtained from GTA.a that received the VHCB
`construct (termed GTA.e), 13 from GTA.b that received
`VHCa (termed GTA.f), 10 from GTA.c that received VLCi3
`(termed GTA.g), and 18 from GTA.d that received VLCa
`(termed GTA.h).
`To determine which of the transfectants expressed VH
`and/or VL chimeric gene segments, we analyzed their RNA
`transcripts by Northern blot hybridization analysis using
`probes specific for the Sp6 VH (Fig. 2A, upper panel) and VL
`(Fig. 2A, lower panel) regions. As a positive control, total
`cytoplasmic RNA from the Sp6 hybridoma gave a strong
`signal corresponding to mRNA of u H chain and K L chain.
`No hybridization was observed with RNA of the untrans-
`fected cells (second lane). The transfectants that received
`VL-containing chimeric genes (represented by a.2 and b.2 in
`
`A
`
`IT ~
`0
`o°d)(D D ao -' GO
`'°L 6 cjajj M
`) 2 o .0
`0 D
`13 a,
`
`U a)
`
`CU
`
`0
`
`V.
`
`V.
`
`qm"4"0 40 ---
`I
`
`VH SP6
`
`so
`
`9W__,_ VL Sp6
`
`to
`
`C) N C%
`2.
`
`C-4 cn
`
`0
`0
`.-.-
`
`CL
`
`aN NcLO
`q O
`-o
`_wC'
`
`\
`
`W.
`
`%e"
`
`_
`
`28S -
`18S-3
`
`28S
`18S --
`
`B k
`
`Da
`150 -
`94 -
`67 -
`43 -
`
`30
`
`Expression of the chimeric TcR chains. (A) Northern blot
`FIG. 2.
`analysis of cytoplasmic RNA using 32P-labeled VHSP6 and VLSP6
`probes. (B) Western immunoblot analysis using anti-Sp6 idiotypic
`mAb and 125I-labeled anti-mouse immunoglobulin. Cell lysates were
`separated by NaDodSO4/PAGE under nonreducing conditions. The
`right panel represents a 4-fold longer exposure of the first five left
`lanes. Transfectants analyzed: a, VLCa; b, VLCI3; C, VHCa; d,
`VHCP; e and h, VLCa + VHCP; f and g, VLC,8 + VHCa.
`
`IL-2 PRODUCTION (U/mi)
`N
`N
`-o
`a)
`C0
`CT
`C
`-~~~~~~~~~ I
`
`-
`
`o,, o
`
`C0
`
`o
`
`TNP-A20-
`
`_____________________
`
`/
`
`Production of IL-2 by transfectants following stimulation
`FIG. 3.
`with EL4 and TNP-modified A.20 cells.
`
`Downloaded from https://www.pnas.org by 136.226.53.3 on February 5, 2023 from IP address 136.226.53.3.
`
`UPenn Ex. 2060
`Miltenyi v. UPenn
`IPR2022-00855
`
`

`

`Immunology: Gross et al.
`
`Proc. Natl. Acad. Sci. USA 86 (1989)
`
`10027
`
`Table 1.
`
`Transfectants expressing chimeric TcR respond to TNP-coupled cells and proteins
`IL-2 production following stimulation, units/ml
`Transfected
`chimeric
`Cell
`TNP-ARHt
`line
`TNP-FyG
`TNP-Spc*
`TNP-A20
`gene(s)
`A20
`EL4
`TNP-KLH
`-
`MD.45
`0
`17
`0
`0
`0
`0
`0
`VLCa
`GTA.a
`26
`0
`1
`0
`0
`0
`3
`VLCI8
`0
`GTA.b
`0
`0
`50
`1
`1
`0
`VHCa
`GTA.c
`71
`0
`30
`0
`0
`1
`11
`VHC,8
`GTA.d
`2
`23
`0
`15
`38
`5
`6
`VHCa + VLCI3
`GTA.g
`8
`33
`0
`2
`136
`84
`9
`VHCI3 + VLCa
`GTA.h
`108
`72
`320
`20
`0
`10
`150
`Transfectants were stimulated with x-irradiated target cells at a 1:1 ratio or with 2 ,ug of TNP-coupled proteins per well
`adsorbed to the wells of microtiter plate. KLH, keyhole limpet hemocyanin; FyG, fowl gamma globulin.
`*Spleen cells of SJL/J mice.
`tHuman lymphoblastoid cell line.
`shown) or when MD.45 hybridoma cells or transfectants that
`expressed only the chimeric VLCa (or CB) constructs were
`used as responders (GTA.a and GTA.b, Fig. 3). These cells
`were able to produce IL-2, as was evident by their ability to
`respond to EL-4 (the H-2b target cell of the MD.45 hybrid-
`oma). Interestingly, transfectants that received and ex-
`pressed the VHCa (GTA.c) or VHCB (GTA.d) chimeric genes
`could be stimulated by TNP-A.20 cells, suggesting that the
`TNP-binding capacity, like the Sp6 idiotype, resides primar-
`ily within the Sp6 VH domain.
`The cTcR recognized TNP in a non-MHC-restricted man-
`ner, as evident by the ability of the transfectants to respond
`to TNP-modified allogeneic and xenogeneic cells (Table 1).
`Transfectants bearing the chimeric TcR could be activated by
`TNP groups bound directly to stimulator cells (Fig. 3, Table
`1) or coupled to a protein carrier (such as TNP-bovine serum
`albumin) and presented either by antigen-presenting cells
`(APCs) such as A.20 and spleen cells (not shown) or adsorbed
`onto plastic substrate (Table 1). TNP-modified proteins in
`solution inhibited specifically and in a dose-dependent man-
`ner the activation of the various transfectants by immobilized
`antigen (Fig. 4A), indicating that, like antibodies, the cTcR
`was able to bind soluble antigen. Anti-SP6-idiotype as well as
`anti-TNP mAbs also inhibited completely the IL-2 produc-
`tion by the cells (Fig. 4 B and C).
`Another manifestation of the functional expression of the
`chimeric receptor, in the effector phase of T-cell response, is
`demonstrated by the ability ofthe transfectants to specifically
`kill haptenated target cells as measured by the 5"Cr-release
`assay. As shown in Fig. 5, all cells studied (except GTA.g2,
`which lost its ability to recognize EL-4 cells) killed EL-4 as
`well as TNP-EL4 cells, and their lytic ability was increased
`following prestimulation with EL-4. However, only the trans-
`fectants could kill the TNP-A.20 target cells. Accordingly,
`stimulation with TNP-A.20 cells enhanced only the reactivity
`
`of cells that expressed the chimeric TcR. Similar results have
`been obtained when TNP-438, H-2s B-lymphoma cells were
`used as targets in the killing assay (data not shown). These
`studies are compatible with the idea that the chimeric recep-
`tor can mediate non-MHC-restricted, antigen-specific target
`cell lysis.
`
`DISCUSSION
`In this article we have shown that chimeric TcR chains,
`composed of immunoglobulin V regions and the TcR C re-
`gions, can be functionally expressed in T cells. Expression of
`this cTcR endowed the recipient T cell with antibody-type
`specificity: it could recognize and respond to stimulator and
`target cells that bear the TNP haptenic group on their surfaces,
`as evident by either IL-2 production (Table 1, Fig. 3) or
`cytolytic activity (Fig. 5). The recognition ofTNP is non-MHC
`restricted, as manifested by the ability of recipients of the
`chimeric genes to respond to TNP-modified cells of different
`strains and species and to TNP adsorbed onto plastic, and is
`independent of the endogenous TcR of the recipient hybrid-
`oma. -The antibody-derived origin of the cTcR is further
`supported by the observation that the anti-Sp6 mAb 20.5
`reacted with the chimeric receptor in immunoblotting (Fig. 2B)
`and immunoprecipitation (24) and completely inhibited the
`activity of most of the transfectants toward TNP (Fig. 4).
`Finally, like their parental antibody, Sp6, the cTcRs also bind
`TNP-modified antigens in solution (Fig. 4A). The ability of
`soluble antigens to inhibit the stimulation of the transfectants
`provides us with direct methods to study the effect of physi-
`cochemical parameters, such as receptor affinity, and anti-
`genic valency, density, and rigidity on T-cell activation.
`The Sp6 antibody-binding site is expressed on the trans-
`fectants regardless of which immunoglobulin V region was
`combined with which TcR C region domains. Both VHCa and
`VLC/8 or VHCP and VLCa chain combinations could pair,
`
`will
`
`B
`{B
`
`A
`
`120
`
`100,
`80
`60
`
`1000
`
`.01
`
`.1
`
`1
`
`1 0
`100
`TNP-BSA (gg/ml)
`Inhibition of IL-2 produced by GTA.g (o) and GTA.h (n) transfectants following stimulation with TNP-A.20 cells. Different
`FIG. 4.
`concentrations of soluble TNP5-BSA (BSA, bovine serum albumin) (A), anti-Sp6 idiotypic mAb (20.5) (B), and anti-TNP mAb (Sp6) (C) were
`added to the reaction wells before adding the responding "transfectomas." (B and C) Culture supernatant containing the mAb.
`
`1 0
`.01
`.1
`Antibody concentration (%)
`
`Downloaded from https://www.pnas.org by 136.226.53.3 on February 5, 2023 from IP address 136.226.53.3.
`
`UPenn Ex. 2060
`Miltenyi v. UPenn
`IPR2022-00855
`
`

`

`10028
`
`Immunology: Gross et A
`
`Proc. Natl. Acad. Sci. USA 86 (1989)
`
`Cell Line
`
`Target Cell Killing Following Stimulation With:
`I
`C7l
`TKID_ A On
`KI-n.
`Id
`
`MD.45
`
`GTA.c2
`
`GTA.d9
`
`GTA.e20
`
`GTA.g2
`
`LWJ
`JWWI L
`II
`20 40 60 80
`20 40 60 80
`20 40 60 80
`51Cr Release(%) from:EL4- O;TNP-EL4- X JNP-A20-E
`
`Cytotoxic activity of transfectants. The different trans-
`FIG. 5.
`fectants and MD.45 parental cells were incubated at an effector:tar-
`get ratio of 2:1 with 51Cr-labeled target cells. Effector cells were used
`either without stimulation or following preincubation with irradiated
`EL4 or TNP-A.20 stimulator cells.
`
`form sulfhydryl bonds, assemble with T3 molecules (ref. 25
`and data not shown), and be expressed in the membrane of
`T cells as functional dimers that bind the hapten, display the
`Sp6 idiotope, and transmit a signal for T-cell activation. In
`most of the transfectants that received both chimeric genes,
`a tetrameric complex of about 160 kDa was also observed
`(Fig. 2B); however, its functional significance has yet to be
`established. Interestingly, in the experimental system we
`employed, using the Sp6 V domains, it appears that the VH
`by itself can account for most of the antibody-binding ca-
`pacity and expression of the idiotypic determinant. Hence,
`transfectants receiving either VHCa or VHCP chimeric genes
`alone expressed functional receptors that bind TNP (GTA.c,
`GTA.d, Fig. 3, Table 1) and react with the anti-Sp6 idiotypic
`antibody 20.5 (Figs. 2B and 4).
`It seems likely that the
`VH-containing cTcR chains can pair with the complementary
`chain of the endogeneous TcR to yield a functional het-
`erodimeric receptor. This was directly shown by the ability
`of anti-VP mAb to immunoprecipitate cTcRs that express
`both VH and the endogenous ,B chain (24). The fact that the
`H-chain V region contributes most of the contact residues in
`the binding site of several antibodies was established by
`studies that assigned antibody activity to isolated chains (26).
`The structural similarity between the TcR Va and VB to the
`immunoglobulin VL, as manifested by the strong resemblance
`in their invariant amino acids (2) and distance between the
`cysteines that form the intrachain disulfide bond (23), may
`explain the efficiency by which the VH chimeric chain pairs
`with either one of the endogenous TcRa or
`chains to yield
`an active binding site.
`It is noteworthy that both quantitative and qualitative
`differences have been observed in the degree of reactivity
`and the ability of the different transfectants that received
`various combinations of chimeric genes to respond to differ-
`ent forms of antigen and stimulator cells. For example, the
`GTA.g series that was first transfected with VHCa and then
`with VLCB responded better to TNP-bovine serum albumin
`than did the GTA.f series that received the same chimeric
`gene combination in reverse order. The GTA.d series that
`was transfected only with VHCB could be better inhibited by
`the 20.5 mAb than the GTA.c series that received the VHCa
`gene. We believe that these differences are due to the level
`of expression of the chimeric TcR and the degree of pairing
`with the endogenous TcR chains. It is likely that pairing of the
`original VH and VL will result in the binding site with the
`highest affinity and specificity toward the hapten.
`Our successful attempts to express functional chimeric
`TcRs that recognize antigen in a non-MHC-restricted manner
`
`pave the way for an approach for the design at will of TcRs
`of any desired specificity, provided that such specificity can
`be predefined by a mAb. Our ability to combine antibody
`specificity with T-cell-mediated target cell lysis may have
`clinical potential: it enables the construction of chimeric TcR
`genes using the V regions of antibodies directed at desired
`antigens on a given target cell. These chimeric genes, once
`produced, are non-MHC-restricted and universal in the sense
`that a given set of chimeric genes could then be transfected
`into T cells from any individual. Upon returning the cells to
`their donors, they should manifest the specificity of the cTcR
`by proliferating and mediating specific effector function
`(cytolysis, production of lymphokines, help, or suppression)
`when encountering their target cells. This approach can be
`exploited, for example, to direct cytotoxic T lymphocytes to
`kill tumor or virally infected cells. Construction of cTcRs
`with anti-tumor specificity will enable testing ofthe feasibility
`of this approach in combating human tumors.
`We are grateful to Dr. G. Kohler and Dr. A. Miyajima for providing
`us with cloned genes and expression vectors, to Dr. S. Schwarzbaum
`for her critical review of the manuscript, and to Ms. M. Fagan for
`excellent secretarial skills. This study is supported by the Ilana
`Hareli Fund from the Israeli Ministry of Health, the Israel Cancer
`Association, and the Leo and Julia Forchheimer Center for Molec-
`ular Genetics. Z.E. is an Incumbent of the Rennete and Marshall
`Ezralow Chair in Chemical and Cellular Immunology.
`Schwartz, R. H. (1985) Annu. Rev. Immunol. 3, 237-261.
`1.
`2.
`Kronenberg, M., Siu, G., Hood, L. H. & Shastri, N. (1986) Annu.
`Rev. Immunol. 4, 529-591.
`Colman, P. M., Deisehofer, J. & Huber, R. (1976) J. Mol. Biol. 100,
`257-282.
`Bear, R., Chen, K.-C., Smith, S. D. & Rabbitts, T. H. (1985) Cell
`43, 705-713.
`Denny, C. T., Yoshikai, Y., Mak, T. W., Smith, S. D., Hollis,
`G. F. & Kirsch, I. R. (1986) Nature (London) 320, 549-551.
`Bear, R., Forster, A. & Rabbitts, T. H. (1987) Cell 50, 97-105.
`Traunecker, A., Dolder, B. & Karijalainen, K. (1986) Eur. J.
`Immunol. 16, 851-854.
`8. -Gascoigne, N. R. J., Goodnow, C. C., Dudzik, K. I., Oi, V. T. &
`Davis, M. M. (1987) Proc. Natl. Acad. Sci. USA 84, 2936-2940.
`Traunecker, A., Dolder, B., Oliveri, F. & Kardalainen, K. (1989)
`9.
`Immunol. Today 10, 29-32.
`Hawley, R. G., Shulman, M. J., Murialdo, H., Gibson, D. M. &
`Hozumi, N. (1982) Proc. Natl. Acad. Sci. USA 79, 7425-7429.
`Ochi, A., Hawley, R. G., Hawley, T., Shulman, M. J., Traunecker,
`A., Kohler, G. & Hozumi, N. (1983) Proc. Natl. Acad. Sci. USA 80,
`6351-6355.
`Mory, Y. Y., Keshet, E., Ram, D. & Kaminchik, Y. (1980) Mol.
`Biol. Rep. 6, 203-208.
`Southern, P. J. & Berg, P. (1982) J. Mol. Appl. Genet. 1, 327-341.
`Mulligan, R. C. & Berg, P. (1981) Proc. Natl. Acad. Sci. USA 78,
`2072-2076.
`Kaufmann, Y., Berke, G. & Eshhar, Z. (1981) Proc. NatI. Acad.
`Sci. USA 78, 2502-2506.
`Ochi, A., Hawley, G. R., Shulman, M. J. & Hozumi, N. (1983)
`Nature (London) 302, 340-342.
`Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular
`Cloning, A Laboratory Manual (Cold Spring Harbor Laboratory,
`Cold Spring Harbor, NY).
`Laemmli, U. K. (1970) Nature (London) 227, 680-685.
`Towbin, H., Staehelin, T. & Gordon, J. (1979) Proc. Natl. Acad.
`Sci. USA 76, 4350-4354.
`Rusconi, S. & Kohler, G. (1985) Nature (London) 314, 330-334.
`Mosmann, T. (1983) J. Immunol. Methods 65, 55-63.
`Schmidt, A.-M., Pettinelli, C. B., Henkart, P. A., Lunney, J. K. &
`Shearer, G. M. (1978) J. Exp. Med. 147, 352-368.
`Becker, D. M., Patten, P., Chien, Y.-H., Yokota, T., Eshhar, Z.,
`Giedlin, M., Gascoigne, N. R. J., Goodnow, C., Wolf, R., Arai,
`K.-I. & Davis, M. M. (1985) Nature (London) 317, 430-434.
`Gross, G., Waks, T. & Eshhar, Z. (1989) in Cellular Basis of
`Immune Modulation, eds. Kaplan, J. G., Green, D. R. & Bleackley,
`R. C. (Liss, New York), pp. 97-101.
`Gross, G., Gorochov, G., Waks, T. & Eshhar, Z. (1989) Transplant.
`Proc. 21, 127-130.
`Painter, R. G., Sage, H. J. & Tanford, C. (1972) Biochemistry 11,
`1327-1337.
`
`18.
`19.
`
`20.
`21.
`22.
`
`23.
`
`24.
`
`25.
`
`26.
`
`3.
`
`4.
`
`5.
`
`6.
`7.
`
`10.
`
`11.
`
`12.
`
`13.
`14.
`
`15.
`
`16.
`
`17.
`
`Downloaded from https://www.pnas.org by 136.226.53.3 on February 5, 2023 from IP address 136.226.53.3.
`
`UPenn Ex. 2060
`Miltenyi v. UPenn
`IPR2022-00855
`
`