A Novel Murine T Cell Receptor Targeting NY-ESO-1

Abstract

Cancer testis (CT) antigens, such as NY-ESO-1, are expressed in a variety of prevalent tumors and represent potential targets for TCR gene therapy. DNA encoding a murine anti-NY-ESO-1 TCR gene (mTCR) was isolated from immunized HLA-A*0201 transgenic mice and inserted into a γretroviral vector. Two mTCR vectors were produced and used to transduce human PBL. Transduced cells were co-cultured with tumor target cell lines and T2 cells pulsed with the NY-ESO-1 peptide, and assayed for cytokine release and cell lysis activity. The most active TCR construct was selected for production of a master cell bank for clinical use. mTCR transduced PBL maintained TCR expression in short-term and long-term culture, ranging from 50–90% efficiency 7–11 days post-stimulation and 46–82% 10–20 days post re-stimulation. High levels of interferon-γ secretion were observed (1000–12000 pg/ml), in tumor co-culture assays and recognition of peptide pulsed cells was observed at 0.1 ng/ml, suggesting that the new mTCR had high avidity for antigen recognition. mTCR transduced T cells also specifically lysed human tumor targets. In all assays, the mTCR was equivalent or better than the comparable human TCR. As the functional activity of TCR transduced cells may be affected by the formation of mixed dimers, mTCRs, which are less likely to form mixed dimers with endogenous hTCRs, may be more effective in vivo. This new mTCR targeted to NY-ESO-1 represents a novel potential therapeutic option for adoptive cell transfer therapy for a variety of malignancies.

Introduction

The adoptive transfer of tumor-infiltrating lymphocytes (TIL) has shown response rates greater than 50% when given in conjunction with IL2 and a lymphodepleting conditioning regimen¹. However, this therapy requires a surgical procedure to procure tumors from which to isolate TIL, and this process results in the generation of TIL in only approximately 70% of resected samples². Recently, alternative methods to engineer lymphocytes to target various tumor antigens were developed, including utilizing gene transfer technology to express new T-cell receptors (TCRs) into peripheral blood lymphocytes.

The NY-ESO-1 protein (gene name CTAG1B), was initially identified by screening a cDNA expression library with an antiserum from a patient with esophageal squamous cell carcinoma, and this tumor antigen is expressed in patients bearing a wide variety of malignancies³. Expression of NY-ESO-1 protein has been observed in approximately one third of melanoma, breast, prostate, lung, ovarian, thyroid and bladder cancer, but its expression in normal tissues is limited to germ cells and trophoblasts⁴. It is thus classified as a cancer testis (CT) antigen. Further studies resulted in the identification of an antigenic peptide corresponding to amino acids 157 to 165 (SLLMWITQC) of the NY-ESO-1 protein as a dominant epitope recognized by HLA-A2 restricted, NY-ESO-1 reactive T cells⁵. This peptide epitope is also shared with the similar CT antigen, LAGE-1. NY-ESO-1 is known to be one of the most immunogenic of the CT antigens, and has therefore been the target of multiple clinical trials, including our previous TCR gene therapy trial, which showed objective clinical responses in four of six patients with synovial cell sarcoma and five of eleven patients with metastatic melanoma⁶.

By introducing a TCR targeting the NY-ESO-1 antigen, large numbers of T cells with defined antigen specificity can be obtained from any patient. While we have observed complete responses in some patient treatments, several short duration partial responses were observed, suggesting that methods to increase the potential efficacy of the TCR were needed. When an exogenous TCR is introduced into a T cell, mixed TCR dimers of the introduced and endogenous TCRs can be formed, which may decrease activity or could lead to non-specific reactivity⁷. Studies in mice have shown that the pairing of endogenous and introduced TCR chains in TCR gene-modified T cells can lead to the formation of self-reactive TCRs, leading to lethal cytokine-driven autoimmune pathology⁸. Murine-human hybrid TCRs have been investigated by constructing TCRs with a murine constant region in place of the human constant region⁹. This resulted in a higher expression of the receptor on the surface of the human lymphocytes, caused by the preferential pairing of the murine constant regions. The chimeric murine-human TCRs mediated higher levels of cytokine secretion and cell lysis, which was likely the result of improved CD3 stability⁹.

While NY-ESO-1 is expressed on a variety of tumors, the levels of antigen expression may be below what is necessary for recognition. Therefore, mechanisms for increasing or inducing the expression of this antigen have been explored. It has previously been shown that the DNA-demethylating agent 5-aza-2’-deoxycytidine (decitabine; DAC) could induce expression of various cancer testis antigens in lung cancer cells, but not normal tissue¹⁰. This drug could potentially be used to up-regulate NY-ESO-1 expression in patients receiving TCR gene modified cells. In this report we describe the development of a novel murine TCR, which recognizes the human NY-ESO-1 antigen. This TCR was designed for use in clinical trials in patients with NY-ESO-1 expressing cancers.

Materials and Methods

Patient peripheral blood mononuclear cells, cell lines, and drug treatment conditions.

All of the peripheral blood mononuclear cells (PBMC) used for transduction and as feeder cells were obtained by apheresis of patients on IRB-approved protocols at the Surgery Branch, National Cancer Institute (Bethesda, MD) and cultured in AIM-V medium (Invitrogen, Carlsbad, CA) supplemented with 5% human AB serum (Valley Biomedical, Winchester, VA) 50 units/mL penicillin (Invitrogen), 50 μg/mL streptomycin (Invitrogen) and 300 IU/mL interleukin-2 (IL-2, Aldesleukin Proleukin®, Novartis, Basel, Switzerland), 25mmol/L HEPES (Invitrogen) and maintained at 37°C with 5% CO₂. Murine lymphocytes were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, 2 mmol/L L-glutamine, 50 units/mL penicillin, 50 μg/mL streptomycin and 30 IU/mL rhIL-2 (R10) (Invitrogen). T2 is a lymphoblastoid cell line deficient in TAP function, whose HLA class I proteins can be easily loaded with exogenous peptides¹¹ and was cultured in R10. Melanoma lines 1300mel (NY-ESO-1+, HLA-A2+), 624.38mel (NY-ESO-1+, HLA-A2+), A375mel (NY-ESO-1+, HLA-A2+), 938mel (NY-ESO-1+, HLA-A2-), 888mel (NY-ESO-1-, HLA-A2-), SK23mel (NY-ESO-1-, HLA-A2+), 1359mel (NY-ESO-1+, HLA-A2-), 1359-A2mel (NY-ESO-1+, HLA-A2+), 624mel (NY-ESO-1+, HLA-A2+), and 1390mel (NY-ESO-1+, HLA-A2+), were generated at the Surgery Branch from resected tumor lesions, as previously described¹² and were cultured in R10 medium. Other cell lines used included: the osteosarcoma cell line Saos2 (NY-ESO-1+, HLA-A2+), (ATCC® HTB-85), the colorectal cancer cell line SW-480 (NY-ESO-1-, HLA-A2+) (ATCC® CCL-228), and the non-small cell lung cancer cell line H1299A2 (NY-ESO-1+, HLA-A2+), which was obtained from the laboratory of Dr. David Schrump (Surgery Branch, National Institutes of Health, Bethesda, MD). The breast carcinoma cell line MDA-MB-435S-A2 (NY-ESO-1+, HLA-A2+), (ATCC® HTB-129), the neuroblastoma cell line SK NAS-A2 (NY-ESO-1+, HLA-A2+), (ATCC® CRL-2137), and the prostate cancer cell line pC3A2 (NY-ESO-1 -, HLA-A2+), (ATCC® CRL-1435), all cultured in R10 media, were transduced with retroviral construct to express HLA-A*0201 as previously described^13,14. COS-A2-ESO, which was transduced with a retroviral vector expressing the NY-ESO-1 gene, and COS-A2-CEA, which was transduced with a retroviral vector expressing the CEA gene were also used and cultured in R10 media.

Deoxyazacytidine (DAC, Sigma Chemical Company, St Louis, MO) treatment was performed by cell exposure to 0.1 μM DAC, 0.5 μM DAC, 1.0 μM DAC, or 10 μM DAC for 72 hours.

Immunization of HLA-A*0201 transgenic mice and isolation of TCR.

Transgenic mice expressing full-length human HLA-A*0201 gene were obtained from the Jackson Laboratory (Bar Harbor, ME). Mouse studies were conducted according to the protocols approved by the National Cancer Institute Animal Care and Use Committee as described previously¹⁴. To generate high avidity mTCRs against NY-ESO-1 CT antigen, eight to twelve week old mice were immunized subcutaneously at the base of the tail with 100 μg of a previously identified naturally processed and presented HLA-A*0201 restricted peptide from NY-ESO-1 (SLLMWITQC) plus 120 μg of helper peptide (HBVc:128–140), emulsified in 100μL of incomplete Freund’s adjuvant (IFA, Montanide ISA 51, Sigma-Aldrich®, St. Louis, MO). Following two immunizations, mice were euthanized and splenocytes were harvested and stimulated in vitro with irradiated T2 cells (18,000 rad) loaded with 1μg/mL of the immunizing peptide in R10 medium containing 30 IU/ml rhIL-2 or LPS-activated HLA-A2+ splenocytes (3,000 rads) pulsed with 1μg/mL priming peptide. Cultures were set up in 24 well plates with 1–3 million splenocytes and 0.2–0.4 million peptide-loaded T2 cells. One week following stimulation bulk murine T-cell cultures were tested in co-culture assays for peptide specific reactivity using LPS-blast cells and T2 cells both pulsed with the relevant peptide after three in vitro stimulations and tumor cell recognition using 888mel, SK23mel, 1359mel, 1359-A2mel, A375mel, 624 mel and 1390mel cells. Antigen specific IFN-γ secretion was measured by ELISA (Thermo Scientific, Rockford, IL). Peptide-reactive bulk cultures were cloned at 10 cells per well in U-bottom 96-well plates, with 5 × 10⁴ peptide-pulsed irradiated T2 cells and 5 × 10⁴ irradiated (3,000 rad) C57BL/6 feeder splenocytes in medium containing 30 IU/mL recombinant human IL-2. Reactive T cells from positive wells from both the first and the third bulk stimulations were then cloned by limiting dilution, and were evaluated for specific recognition of peptide and tumor cells by means of specific IFN-γ secretion.

Cloning of NY-ESO-1 specific, HLA-A*0201 restricted TCR.

Total RNA was extracted from tumor reactive T cell clones using RNeasy mini kit (Qiagen, Valencia, CA). TCR α and β-chains from each tumor reactive T cell clone were cloned using SMART™ RACE cDNA amplification kit (Clontec, Mountain View, CA) with gene specific primers in the constant region of mouse TCR α and β-chains¹⁴. After the identification of the variable regions of α and β-chains and the specific constant region of the β-chain, specific primers were used to amplify the full length TCR α and β-chains from the cDNA. For the amplification of TCRs gene specific primers were made from the constant region of mouse TCR α and β-chains. The PCR products of the 5’-RACE were cloned into PCR2.1 TOPO vector (Invitrogen Life Technologies) and the insert DNA fragments were sequenced. The DNA sequence data were analyzed using The International Immunogenitics Information System® (http://imgt.cines.fr/IMGT_vquest/vquest?livret=0&Option=mouseTcR) for the identification of mouse TCR α and β-chains. Following the identification of the variable regions of α and β-chains and the identification of constant region of the β-chain (CB1 or CB2), specific primers were used to amplify the full length TCR α and β-chains from the cDNA.

Genetic modification of T lymphocytes via RNA electroporation.

Transient transfection of T lymphocytes with murine TCR α and β chains was done by electroporation with in vitro transcribed (IVT) RNA as previously described^15,16. Whole PBMCs were stimulated with 30 ng/mL OKT3 (MuromonAB-CD3, Orthoclone OKT3, Ortho Biotech, Raritan, NJ) and 300 IU/mL rhIL-2. For some experiments, CD8+ and/or CD4+ T cells were purified 2 to 4 days later using anti-CD8 and/or anti-CD4 coated magnetic beads (Miltenyi Biotec). Five to seven days after stimulation, T cells were washed and gently resuspended in Opti-MEM (Invitrogen) at 2 × 10⁷/mL. Afterwards, 0.05 to 0.2 mL cells were mixed with IVT RNA (2 μg each of α and β chain encoding RNA per 1 × 10⁶ T cells) and electroporated with a single pulse in a 2 mm cuvette (Harvard Apparatus BTX) using an ECM 830 Electro Square Porator at 500 V and 500 μs (Harvard Apparatus BTX). After electroporation, the cells were transferred to fresh complete medium and incubated at 37° C.

Construction of γ−retroviral vectors expressing NY-ESO-1 specific HLA-A*0201 restricted TCRs.

The TCR α and β chains that showed the highest level of specific reactivity were cloned into the MSGV1 based γ−retroviral vector, after synthesis of codon-optimized sequences (Blue Heron, Bothwell, WA) with the transgene constructs arranged in the following orders of configuration: 1-TCR α chain, linker peptide furinSGSGP2A and TCR β chain and 2-TCR β chain, linker peptide furinSGSGP2A and TCR α chain. The TCR expression in this vector is driven by the viral LTR, α and β chains are expressed as a single open reading frame using the 2A linker peptide. The cloned TCR inserts were verified by restriction enzyme digestion and DNA sequencing.

Transduction of PBL.

γ−Retroviral supernatants were generated by transfecting respective MSGV1-ESO-mTCR vector DNA from each of the constructs with a plasmid encoding RD114 envelope into 293-GP cells using the Lipofectamine 2000 reagent (Invitrogen) in Optimem medium (Invitrogen). Retroviral vector expressing hTCR against NY-ESO-1 (1G4- α 95:LY) was used as a positive control⁶. Retroviral vector expressing TCR against gp100 (154) or green fluorescent protein (GFP) were used as negative controls. Viral supernatants were loaded onto RetroNectin (Takara Bio, Otsu, Shiga, Japan) coated non-tissue culture treated 6 well plates. PBL were stimulated with 50 ng/ml OKT3, and rhIL2 (300 IU/ml) 72h prior to transduction and the transduction was carried out as described previously^13,14. After harvest of one-time stimulated cells for testing (S1d7–14), cells were rapidly expanded (REP) in the presence of soluble OKT3 (300 IU/ml), IL2 (6,000 IU/mL) and irradiated feeders as previously described^6,13. After day 5 of REP, cells were maintained in culture at 0.5–1.0 × 10⁶ cells/mL until harvested for testing on days 7–10 (R2d7–10).

Production of retroviral supernatant from stable producer cell clones.

PG13 packaging cell clones were generated using PG13 gibbon ape leukemia virus packaging cell line (ATCC, Manassas, VA) and human ecotropic package cell line Phoenix Eco (provided by the National Gene Vector Biorepository, Indiana University, Indianapolis, IN). Supernatant from each clone generated was used to transduce human PBL. Reactivity was evaluated by measurement of IFNγ release by ELISA (Thermo Scientific, Rockford, IL) after transduced T cells were co-cultured with target tumor cell lines for 16 hrs.

Flow Cytometry Analysis (FACS)

NY-ESO-1 mTCR transduced T cells were stained with either a FITC or PE-conjugated monoclonal antibody against the constant region of the murine β chain (eBioscience, San Diego, CA) and FITC or PE-conjugated anti-human CD8 antibody (Becton Dickson Immunocytometry Systems, San Jose, CA). PE-conjugated anti-human Vβ13.1 (Beckman Coulter, Miami, FL) was used to detect the NY-ESO-1 hTCR. Cells were analyzed using a FACScan flow cytometer with CellQuest software (BD Biosciences, San Jose, CA) or FlowJo software (Tree Star, Inc, Ashland, OR).

Cytokine release assay.

TCR engineered PBLs were tested for antigen specific reactivity in cytokine release assays using peptide loaded T2 cells or tumor cells. In these assays effector cells (1 × 10⁵) were co-cultured with equal number of target cells in AIM-V medium in a final volume of 0.2 mL in duplicate wells of a 96-well U-bottom microplate. Culture supernatants were harvested 18–24 hours after the initiation of co-culture and assayed for IFN-γ by ELISA (Thermo Scientific).

Cytotoxicity assay.

The ability of the transduced PBL to lyse HLA-A*0201⁺/NY-ESO-1⁺ tumor cells was measured using a CytoTox-Glo™ bioluminescence assay (Promega, Madison, WI). This assay utilizes a luminogenic peptide substrate, the AAF-Glo™ Substrate to measure dead-cell protease activity, which is released from cells that have lost membrane integrity, resulting in the generation of a “glow-type” luminescent signal that is proportional to the number of dead cells in the sample. The AAF-Glo™ Substrate cannot cross the intact membrane of live cells and does not generate any appreciable signal from the live-cell population. In these assays TCR engineered PBL effector cells were co-incubated with increasing ratios of target cells (E:T) in AIM-V medium in 96-well U-bottom plates at 37°C for 4 hr. Lysis was measured by bioluminescence release in the medium: percent specific lysis = [specific release – (spontaneous effector release + spontaneous target release)]/total target release – spontaneous target release × 100%, average of quadruplicate samples.

CD4/CD8 separation.

NY-ESO-1 hTCR and mTCR engineered CD4⁺ and CD8⁺ populations were separated using magnetic beads-based BD IMag™ human CD4 or CD8 T lymphocyte enrichment set-DM kit for negative selection of those subsets (Miltenyi Biotec Inc., Auburn, CA).

Real-time reverse transcription-PCR analysis.

RNA was isolated using RNeasy Mini kit (Qiagen). cDNAs were made using Reverse Transcription Kit (Bio-Rad, Hercules, CA). Quantitative reverse transcription (qRT)-PCR primers for the NY-ESO-1 gene and β-actin expression have been previously described¹⁷.

Immunoblotting.

Detection of protein was determined by Immunoblot analysis as previously described¹⁸. Total cellular proteins were extracted using the RIPA buffer lysis kit (Millipore, Billerica, MA) supplemented with 1X protease inhibitor (Roche, San Francisco, CA), and 1 mmol/L phenylmethylsulfonyl fluoride (Sigma-Aldrich, LLC, St. Louis, MO). Cell lysates were resolved on 4–20% trisglycine gels (Invitrogen), transferred to PVDF membranes (Millipore), and incubated overnight with an anti-NY-ESO-1 primary antibody E978 (Sigma). Immunoblot signals were detected using appropriate HRP-conjugated secondary antibodies (Santa Cruz) and SuperSignal West Pico Chemiluminescent Substrate (Pierce Biotechnology, Rockford, IL).

Results

Generation of NY-ESO-1 reactive murine T cell clones from HLA-A*0201 transgenic mice.

Transgenic mice expressing the full-length HLA-A*0201 molecule were immunized with a previously identified naturally processed and presented HLA-A*0201 restricted peptide from NY-ESO-1. Following two immunizations T cells were harvested from the spleen and stimulated. Following three rounds of in vitro stimulation, reactive T cell wells were cloned by limiting dilution and tested for antigen specific reactivity. Five clones secreted high levels of IFN-γ in response to peptide pulsed T2 cells and HLA-A2+ and NY-ESO-1+ tumor cells and secreted only low levels of IFN-γ in response to tumor cells negative for either HLA-A2 or NY-ESO-1 (Table 1). The α and β genes for the individual clones were isolated and inserted into a RNA expression vector. In vitro transcribed mRNA for the α and β chains was introduced into human T cells by electroporation, and assayed for reactivity by co-culture. High levels of IFN-γ were observed in PBL electroporated with TCR α chain TRAV6D and β chain TRBV26 (Table 2). These specific chains were thus chosen for further study.

Table 1.

Initial screening of bulk populations of murine anti-NY-ESO-1 T cells. NY-ESO-1:157–165: Murine T cell clone reactivities (pg/ml IFNg) shown post 3 bulk stimulations with the indicated cell lines. TCR chain gene usage was as shown.

	T2+ HBV	T2+ESO:157	media	888	SK23	1359	1359-A2	A375	624	1390
HLA-A2	+	+	−	−	+	−	+	+	+	+
NY-ESO-1	-	+	−	−	−	+	+	+	+	+
TCR Usage
TRAV7D-3/TRBV14	317	>14,000	314	316	309	383	6,027	388	1,931	17,567
TRAV6D/TRBV26	223	>14,000	231	208	210	206	2,523	232	2,527	352
TRAV6D/TRBV26	306	>14,000	266	234	278	347	1,477	556	4,629	517
TRAV7D-3/TRBV26	335	>14,000	232	241	208	271	3,661	557	4,487	379
TRAV6D/TRBV26	239	>14,000	222	208	207	229	2,272	737	7,139	716

Table 2.

Screening of 5 murine anti-NY-ESO-1 T cell clones. IFNγ (pg/ml) production by human PBL after RNA electroporation of specific α and β chains co-cultured with various HLA-A2 +/− and NY-ESO-1 +/− cell lines after stimulation with OKT3 and IL-2. TCR chain gene usage was a shown.

		T2+HBV	T2+ESO:157	media	888	SK23	1363	1390	A375	624	COS-A2-CEA	COS-A2-ESO
	HLA-A2	−	−	−	−	+	+	+	+	+	+	+
	NY-ESO-1	−	+	−	−	−	+	+	+	+	−	+
Donor A
	GFP	326	180	8	198	134	220	325	632	32	94	74
	hTCR LY	235	>10,000	7	188	81	>10,000	1637	5752	531	52	2320
	TRAV7D-4/TRBV19	664	>10,000	7	185	102	2262	489	893	39	72	449
	TRAV13D-2/TRBV14	197	>10,000	8	152	88	134	159	338	26	75	55
	TRAV7D-3/TRBV14	155	366	7	129	112	122	171	378	28	68	71
	TRAV6D/TRBV26	198	>10,000	11	255	156	>10,000	3859	9973	782	91	3872
	TRAV7D-3/TRBV26	190	1269	0	208	107	246	189	509	34	79	104
Donor B
	GFP	26	30	2	47	27	22	62	98	8	16	19
	hTCR LY	50	>10,000	4	32	22	2484	208	192	150	11	588
	TRAV7D-4/TRBV19	183	>10,000	2	34	15	149	47	39	7	13	90
	TRAV13D-2/TRBV14	22	7898	9	27	13	20	42	58	11	22	17
	TRAV7D-3/TRBV14	24	23	11	28	13	21	30	39	5	7	13
	TRAV6D/TRBV26	63	>10,000	39	77	40	3597	777	344	133	32	683
	TRAV7D-3/TRBV26	10	77	0	28	17	27	33	51	5	16	40

Construction of NY-ESO-1 mTCR expressing retroviral vector and transduction of PBL.

Two γ−retroviral vectors were constructed to express the mTCRα (TRAV6D) and mTCRβ (TRBV26) chains expressed as a single open reading frame using the 2A linker peptide, with the order of the chains alternated (Fig. 1A). Human PBL were stimulated for 2 days and then transduced with retroviral vector supernatant. FACS analysis of transduced PBL using an anti-mouse TCR-β chain revealed that both CD8⁺ and CD4⁺ cells had been efficiently transduced with the two mTCR vectors (55%−72%, Fig. 1B).

Figure 1.

mTCR vector analysis. A. Schematic illustration of the MSGV1 based retroviral vector encoding anti NY-ESO-1 murine T cell receptor expression cassette. TCR α and β chains are linked with furin-spacer (SGSG)-P2A ribosomal skip peptide sequence. B. Flow cytometric analysis of NY-ESO-1 mTCR transduced PBL, two donors shown, representative of eleven different donors. C. Recognition of peptide pulsed T2 cells by the two different NY-ESO-1 mTCR transduced PBL. Human PBL expressing mTCR against NY-ESO-1, GFP (negative control) and untransduced PBL (negative control) were co-cultured for 16h with T2 cells that were previously pulsed with the indicated concentrations of peptide. Data are representative of six different donors. D. Recognition of NY-ESO-1+, HLA-A2+ tumor cell lines (H1299-A2, 624.38) NY-ESO-1-, HLA-A2- tumor cell line (888) and NY-ESO-1+, HLA-A2- tumor cell line (938). NY-ESO-1 mTCR transduced PBL were co-cultured for 16 hours with tumor target cell lines and IFNγ levels were then measured, representative of 6 different donors.

Evaluation of the function of NY-ESO-1 mTCR vector engineered PBL.

To evaluate the reactivity of the two NY-ESO-1 mTCR constructs (NY-ESO-1 mTCR α/β and NY-ESO-1 mTCR β/α), transduced PBL were co-culture with peptide-pulsed T2 cells. PBL transduced with both mTCR constructs specifically secreted IFN-γ upon encounter with as little as 1ng/mL of the antigenic peptide in a dose-dependent manner (Fig. 1C). Co-culture of PBL expressing the two vectors with T2 cells, produced background levels of IFN-γ. Transduction of a total of five donors indicated that the NY-ESO-1 mTCR β/α generally produced higher levels of IFN-γ secretion when compared to the NY-ESO-1 mTCR α/β for the same level of peptide. To assess the recognition of tumor cells, the mTCR engineered PBL were co-cultured with a panel of HLA-A*0201⁺ and HLA-A*0201⁻ melanoma and lung tumor derived cell lines. Specific release of IFN-γ was observed when the TCR engineered PBL were co-cultured with HLA-A*0201⁺/NY-ESO-1⁺ cell lines but not HLA-A*0201⁻/NY-ESO-1⁺ cell lines (Fig. 1D). In general, T cells transduced with the NY-ESO-1 β/α mTCR released modestly higher levels of IFN-γ in response to HLA-A*0201⁺/NY-ESO-1⁺ tumor cell targets (Fig. 1D). Based on this analysis we selected the mTCR targeting NY-ESO-1: β/α retroviral vector construct for further analysis.

Screening of the PG13 packaging cell clones.

DNA encoding the mTCR NY-ESO-1 β/α vector was used to produce retroviral vector producer cell clones under conditions required for clinical application. Supernatant from six PG13 producer cell clones was used to transduce human PBL. FACS analysis of transduced PBL using the anti-mouse TCR-β chain revealed that each clone produced virus that mediated positive TCR transduction (range 30–63% transduction). A comparison of the six mTCR PG13 producer clones showed that T cells transduced with clone C1 released high levels of IFN-γ specifically in response to HLA-A*0201⁺/NY-ESO-1⁺ tumor cell target H1299-A2 and demonstrated the highest transduction efficiency (Fig. 2). Based on this analysis we selected clone C1 for the production of a master cell bank for subsequent production of good manufacturing practice (GMP) retroviral supernatant.

Figure 2.

Screening of PG13 producer cell clones. Equal volumes of supernatant from six different PG13 cell clones were used to transduced human PBL and these cultures tested for antigen reactivity by co-culture with tumor cell lines. Culture supernatants were harvested 18–24 hours after the initiation of co-culture and assayed for IFN-γ by ELISA. Transduction efficiency was measured by FACS analysis using an anti-mouse TCRβ monoclonal antibody, % of mTCRβ positive cells listed.

Comparison of the NY-ESO-1 mTCR with the NY-ESO-1 hTCR

This new murine-derived anti-NY-ESO-1 mTCR (mESO C1) was then compared to our previously reported human TCR (hESO LY), which recognizes the same peptide epitope. To compare the respective NY-ESO-1 TCRs (murine, or mTCR, versus human, or hTCR), FACS analysis of transduced PBL using the anti-mouse TCR-β chain and the anti-Vβ13.1 antibodies was performed after one stimulation with OKT3 and following a second large-scale expansion using the rapid expansion protocol (REP) (Fig. 3A). Results from four independent donor transductions demonstrated that the mTCR and the hTCR had equivalent percentages of transduction after initial stimulation, with the mTCR showing small but statistically significant increases in its expression efficiency after REP (p=0.043, Fig. 3A). Both the mTCR and the hTCR specifically secreted IFN-γ upon encounter with as little as 0.1 ng/mL of the antigenic peptide in a dose-dependent manner after one stimulation with OKT3 and after REP (Fig. 3B). Following the REP, the mTCR released higher levels of IFN-γ compared to the hTCR vector transduced T cells at each concentration of peptide (Fig. 3B). Co-culture of PBL expressing NY-ESO-1 mTCR or NY-ESO-1 hTCR with control T2 cells that were not pulsed with any peptide, produced background levels of IFN-γ. To assess the specific recognition of tumor cells, the mTCR engineered PBL were co-cultured after REP with a panel of NY-ESO-1⁺/HLA-A*0201⁺ and NY-ESO-1⁺/HLA-A*0201⁻ melanoma and lung tumor derived cell lines. Specific release of IFN-γ was observed when both the mTCR engineered PBL and the hTCR were co-cultured with HLA-A*0201⁺/NY-ESO-1⁺ cell lines but not NY-ESO-1⁺/HLA-A*0201⁻ cell lines (Fig. 4A).

Figure 3.

Comparison of human and murine anti-NY-ESO-1 TCRs. A. Activated human PBL were transduced with equal volumes of hTCR or mTCR vector supernatant and the percentage of TCR positive cells (NY-ESO-1 hTCR vs. NY-ESO-1 mTCR) determined. Data shown are seven days post initial stimulation with OKT3 (S1) and then following a 14-day rapid expansion protocol (R1). Data are representative of four different donors. B. Recognition of peptide pulsed T2 cells by the NY-ESO-1 hTCR and NY-ESO-1 mTCR transduced PBL after S1 and R1. Human PBL expressing hTCR or mTCR against NY-ESO-1, GFP (vector control) and untransduced PBL (negative control) were co-cultured for 16h with T2 cells that were previously pulsed with different concentrations of peptide, representative of four different donors.

Figure 4.

Tumor cell line reactivity. A. Recognition of NY-ESO-1+, A2+ tumor cell lines (H1299-A2, 624.38 and 1300) and NY-ESO-1+, A2- tumor cell line (938). NY-ESO-1 hTCR and NY-ESO-1 mTCR transduced PBL that had undergone REP were co-cultured for 16 hours with tumor target cell lines and IFNγ levels were then measured, representative of 4 different donors. B. Cell lysis assay comparing cell specific lysis activity of the NY-ESO-1 hTCR transduced PBL with NY-ESO-1 mTCR transduced PBL after REP. Effector cells were co-cultured with an NY-ESO-1+, HLA-A2+ tumor target (624.38, + control) and an NY-ESO-1+, HLA-A2- tumor target (938, - control), representative of four different donors.

Both mTCR and hTCR transduced PBL demonstrated similar lytic activity against melanoma NY-ESO-1⁺/HLA-A*0201⁺ tumor cell line 624.38mel (Fig. 4B). There was no lysis of NY-ESO-1⁺/HLA-A*0201⁻ cell line 938 mel, and the GFP transduced PBL showed no reactivity against any of the target cells (Fig. 4B). The anti-tumor antigen reactivity of CD4⁺ T cells following transduction with NY-ESO-1 hTCR and NY-ESO-1 mTCR was determined by purifing T cells with CD4+ magnetic beads, then co-culture for 16 hours with a panel of NY-ESO-1⁺/HLA-A*0201⁺ and NY-ESO-1⁺/HLA-A*0201⁻ melanoma and lung tumor derived cell lines. CD4⁺ T lymphocytes transduced with both the mTCR and the hTCR demonstrated specific release of IFN-γ when co-cultured with NY-ESO-1⁺/HLA-A*0201⁺cell lines but not NY-ESO-1⁺/HLA-A*0201⁻ cell lines (Fig. 5).

Figure 5.

Activity in human CD4+ T cells. NY-ESO-1 hTCR (hESO LY)and NY-ESO-1 mTCR (mESO C1) transduced PBL were bead sorted with CD4+ magnetic beads, then co-cultured for 16 hours with tumor target cell lines (H1299-A2, A375, 624.38, 938) and IFNγ levels were then measured, representative of 4 different donors.

Recognition of multiple tumor histologies.

There are a variety of tumors known to express the NY-ESO-1 antigen. To assess the specific recognition of various tumor histologies, NY-ESO-1 mTCR transduced PBL were co-cultured with various HLA-A*0201⁺/NY-ESO-1⁺ cell lines derived from melanoma, non-small cell lung cancer, neuroblastoma, breast cancer, and osteosarcoma. Specific release of IFN-γ was observed for each histology tested, with levels ranging from >3000 pg/ml, up to 20,000 pg/ml (Fig. 6), with significantly less IFN-γ release when the untransduced cells were co-cultured with the different histologies.

Figure 6.

Recognition of multiple histologies. NY-ESO-1 mTCR transduced PBL were co-cultured for 16 hours with tumor target cell lines of different histologies (H1299-A2=non-small cell lung cancer, A375=melanoma, Saos2=osteosarcoma, MDA-435S-A2= breast adenocarcinoma, SKN AS-A2= neuroblastoma) and IFNγ levels were then measured. Representative of 2–6 donors. UT, untransduced.

Effects of DAC treatment on NY-ESO-1 gene expression.

DAC, a DNA demethylating agent, is known to induce expression of various cancer testis antigens in non small cell lung cancer tumor cell lines but not in normal tissue¹⁷. To extend this observation to other histologies, two cell lines known to be NY-ESO-1⁻ /HLA-A*0201⁺ derived from colorectal cancer (SW480) and prostate cancer (PC3A2) were treated with increasing concentrations of DAC: 0.1 μM, 0.5 μM, 1 μM or 10 μM) for 72 hours. Quantitative PCR (Taqman) analysis showed increasing copy numbers of NY-ESO-1 corresponding to increasing concentrations of DAC for the colorectal cancer cell line, up to 25,000 × greater than the untreated control (p<0.0001) and increasing copy numbers of NY-ESO-1 corresponding to increasing levels of DAC up to the level of 10 μM for the prostate cancer cell line, 160,000 × greater than the untreated control (p<0.0001) (Fig. 7A). On immunoblot assay for protein expression, increasing levels of protein were also demonstrated with increasing DAC concentration for both the colorectal cancer cell lines and the prostate cancer cell line (Fig. 7B). These DAC-treated tumor cells were co-cultured with mTCR transduced PBL for 16 hrs. Supernatant was then collected and IFN-γ release measured. Both histologies (colorectal cancer and prostate cancer) showed escalations of IFN-γ release with increasing concentrations of DAC, with double to quadruple increase of the amount of IFN-γ released for the colorectal cell line and a log increase in the amount of IFN-γ released for the prostate cancer cell line (Fig. 7C). For the prostate cancer cell line the highest DAC concentration resulted in lower IFN-γ in culture, which was correlated with both Taqman and immunoblot assays, and was most likely due to cell toxicity at this dose level (Fig. 7C).

Figure 7.

Epigenetic modification permits NY-ESO-1 TCR recognition. A. Taqman analysis on cells treated with increasing concentrations of DAC for 72 hrs with copy number relative to β-actin control. B. Immunoblot demonstrating increasing protein expression of NY-ESO-1 with increasing concentrations of DAC exposure, shown with β-actin control. C. NY-ESO-1 mTCR (mESO C1) transduced PBL were co-cultured for 16 hours with tumor target cell lines of different histologies that had been exposed to increasing concentrations of DAC for 72 hours (PC3A2=Prostate cancer, SW480= Colorectal cancer) and IFNγ levels were then measured. Representative of two donors. UT, untransduced.

Discussion

The expression of NY-ESO-1 has been demonstrated in cancers such as sarcomas, specifically synovial cell sarcoma and myxoid/round cell liposarcoma¹⁹, triple negative breast cancer²⁰, hepatocellular carcinoma²¹, urothelial carcinoma²², and non-small cell lung cancer²³, among multiple other malignancies, but displays restricted expression in normal tissues. To date, there has been one reported clinical trial utilizing anti-NY-ESO-1 human TCR transduced cells to target cancer, and responses were seen in patients with melanoma and synovial cell sarcoma⁶. Given our prior results in targeting the NY-ESO-1 CT antigen with human TCR transduced cells in this clinical trial, we sought to identify means to enhance the functionality of TCR-transduced cells.

The functional activity of TCR transduced cells may be affected by the formation of mixed dimers between endogenous and transduced TCR chains, which would result in the dilution of the introduced TCR, potentially translating into poorer cellular activity⁹. Multiple studies have been conducted examining ways of improving TCR function through genetic modifications^9,24,25. mTCRs have been shown to be less likely to form mixed dimers with endogenous hTCRs, and thus may be more effective in vivo. Goff et al demonstrated that by substituting the human constant region with murine constant regions, increased receptor expression and tetramer binding, as well as, increased in vitro antitumor activity of Mart-1 TCR transduced cells was obtained following a rapid expansion protocol²⁵.

In the comparisons reported herein, the hTCR, and mTCR generally had equal transduction efficiency, (as measured by FACS analysis), with the mTCR demonstrating superior TCR expression post-REP. The mTCR was found to be a highly avid receptor and displayed antigen specific recognition of both peptide-pulsed T2 cells and tumor target cell lines. The mTCR also demonstrated equivalent cell lysis, and in subset analysis, isolated CD4⁺ cells maintained anti-tumor activity against target tumor cell lines. Of note, the natural hTCR required modification of amino acids within the CDR3 region to possess similar CD4+ cell anti-tumor activity against target tumor cell lines²⁶.

While multiple histologies (melanoma, non-small cell lung cancer, breast cancer, osteosarcoma, neuroblastoma) are recognized by the mTCR, using DAC exposure, we demonstrated that NY-ESO-1 expression can be induced in additional histologies, specifically colorectal cancer and prostate cancer, with subsequent recognition by mTCR transduced cells. This illustrates the potential for a wide array of cancer histologies that could be future targets for this mTCR. One such example is neuroblastoma, a disease in which approximately half of patients with stage IV disease will relapse despite current therapy, with very little likelihood of achieving a cure²².

Prior publications have demonstrated the upregulation of NY-ESO-1 on neuroblastoma cells after exposure to DAC²⁸. Krishnadas et al reported on a pediatric patient with relapsed neuroblastoma who was treated on a phase I study using DAC and a dendritic cell vaccine targeting NY-ESO-1 resulted in complete remission 1 year from their last treatment²⁸. The expression of NY-ESO-1 has been shown to be inducible in chondrosarcomas following treatment with DAC, and these tumors were then recognized by NY-ESO-1 specific CD8⁺ effector cells²⁹. In addition, DAC has been shown to derepress NY-ESO-1 expression on acute myeloid leukemia cell lines, which were then recognized by CD8⁺ T-cell clones³⁰. These examples demonstrates additional possibilities to expand the use of our mTCR to target tumors that do not innately express NY-ESO-1, or that do not express the antigen at levels that are high enough for recognition through upregulation of antigen expression by exposure to DAC.

Although to date it has not been observed in humans, there has been concern over the potential for autoimmune pathology in the utilization of TCR gene therapy. Bendle et al demonstrated lethal GVHD in mouse models of TCR gene therapy⁸ which, in theory, might be lessened using mTCRs in transduced human T cells. We have previously implemented three human clinical trials using murine TCR transduced cells targeting the antigens MAGE-A3, CEA, and gp100, the results of which have been previously reported^13,31,32. While anti-tumor responses using these mTCRs varied, to date there have been no cases of GVHD in any of these, or other TCR gene therapy clinical trials. One drawback in the use of mTCRs may be the development of human anti-mTCR immune responses. Davis et al reported that while a subset of patients treated with murine T cell receptors developed antibodies directed against the mTCR variable regions, there was no association of the development of a host immune response towards mTCR transduced cells and the clinical outcome³³.

The pre-clinical data presented herein demonstrated the increased functionality of a new mTCR targeting NY-ESO-1, and suggest that this mTCR could be used in clinical trials to treat patients with a variety of cancer histologies.