Abstract
T cell receptor-engineered T cells (TCR-Ts) therapy is promising for cancer immunotherapy. Most studies have focused on identifying tumor-specific T cell receptors (TCRs) through predicted tumor neoantigens. However, current algorithms for predicting tumor neoantigens are unreliable and many neoantigens are derived from non-coding regions. Thus, the technological platform for identifying tumor-specific TCRs using natural antigens expressed on tumor cells is urgently needed. In this study, tumor organoids-enriched tumor infiltrating lymphocytes (oeT) were obtained by repeatedly stimulation of autologous patient-derived organoids (PDO) in vitro. The oeT cells specifically responded to autologous tumor PDO by detecting CD137 expression and the secretion of IFN-γ using enzyme-linked immunospot assay. The measurement of oeT cell-mediated killing of three-dimensional organoids was conducted using a caspase3/7 flow cytometry assay kit. Subsequently, tumor-specific T cells were isolated based on CD137 expression and their TCRs were identified through single-cell RT-PCR analysis. The specificity cytotoxicity of TCRs were confirmed by transferring to primary peripheral blood T cells. The co-culture system proved highly effective in generating CD8+ tumor-specific oeT cells. These oeT cells effectively induced IFN-γ secretion and exhibited specificity in killing autologous tumor organoids, while not eliciting a cytotoxic response against normal organoids. The analysis conducted by TCRs revealed a significant expansion of T cells within a specific subset of TCRs. Subsequently, the TCRs were cloned and transferred to peripheral blood T cells generation engineered TCR-Ts, which adequately recognized and killed tumor cell in a patient-specific manner. The co-culture system provided an approach to generate tumor-specific TCRs from tumor-infiltrating lymphocytes of patients with colorectal cancer, and tumor-specific TCRs can potentially be used for personalized TCR-T therapy.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
With the developing landscape of cancer treatment, immunotherapy has become a major component in the management of multiple cancers. Empirical evidence has demonstrated the efficacy of T cell receptor-engineered T cell therapy in inducing potent anti-tumor responses in several solid tumors through adoptive cell transfer (ACT) [1,2,3,4]. These clinically effective anti-tumor responses are mediate by tumor-specific TCRs recognition of tumor antigens. However, the major bottleneck in the wide application of TCR-T therapy is how to efficiently and accurately isolate tumor-specific TCRs. Most studies have conducted comprehensive tumor-specific TCRs screening by incubating antigen presenting cells (APCs) with T cells, which relied on either loading synthetic neoantigen peptides onto APCs or electroporating tandem minigenes encoding tumor neoantigen into APCs to generate tumor cell surrogates [5, 6]. These methods highly depends on the prediction and using appropriate tumor antigens, while current prediction algorithms for tumor neoantigens are unreliable and some neoantigen peptides can’t be a naturally presented epitope [7, 8]. Furthermore, only a limited number of neoantigens have the potential to elicit an immune response in epithelial cancer [9], and many tumor neoantigens were derived from non-coding regions [10]. Currently, there is an urgent need to find more sources of tumor antigens.
Tumor organoids presenting various natural tumor neoantigens could be used in an unbiased manner to generate tumor-specific T cells [11, 12]. Tumor organoids preserve histopathologic features and genomic characteristics of the original tumor, including individual cancer mutations and mutation burden [13,14,15]. Previous studies have demonstrated that colorectal tumor organoids displayed diverse HLA peptide presentation, including tumor-specific antigens, that renders organoids a suitable model for the generation of natural tumor antigens for stimulating the expansion of tumor-specific T cells in vitro [16].
In this study, tumor organoids are used as the source of antigens to address the issues of imprecise antigen prediction and time-consuming tumor antigen synthesis. As well as for the source of TCR repertoire, we choose autologous tumor-infiltrating lymphocytes (TILs), due to TILs have greatly enriched for tumor specific T cells and are optimal for the identification of tumor-specific TCRs [17,18,19,20]. In addition, enriching tumor-specific T cells in vitro can avoid immunosuppressive interference in the tumor microenvironment through culture optimization.
Here, we developed a co-culture system of tumor-specific T cells that can be efficiently expanded by co-culturing tumor organoids with autologous TILs, and subsequently their TCRs were isolated and transferred to peripheral T cells to generate tumor-specific TCR-Ts. The capacity of TCR-T cells to exhibit tumor-specific reactivity and cytotoxicity against autologous tumor cells have been observed and evaluated. This strategy provides an approach to identify tumor-specific TCRs in vitro rapidly, which can be a potential strategy for future clinical application of personalized TCR-T therapy.
Materials and methods
Patient samples
Tumor tissues and peripheral blood were obtained from patients with a confirmed diagnosis of colorectal cancer. The fresh surgically resected tumor samples were placed in sterile 15 mL tubes containing Advanced DMEM/F12 (GIBCO) supplemented with pencillin/streptomycin (GIBCO), subsequently divided into multiple fragments for TILs culture and generation of patient derived organoids (PDO). This study was approved by the Medical Ethical Committee of the Peking University Shougang Hospital and written informed consent was obtained from all patients.
PDO and TILs establishment
The establishment of PDO was performed in accordance with the previously described method [12]. Briefly, the fresh tumor tissue and normal colon tissue was washed with ice-cold phosphate-buffered saline (PBS) buffer and minced into small fragments. The fragments were enzymatically digested using 10 μg/mL hyaluronidase type IV (Sigma-Aldrich), 1.5 mg/mL collagenase II (Sigma-Aldrich), and 10 μM Y-27632 (Sigma-Aldrich) for approximately 45 min before embedding in Matrigel (Corning). Tumor organoids were cultured in Advanced DMEM/F12 (Gibco) supplemented with 10 mM HEPES (Gibco), 1 × B27 supplement without vitamin A (GIBCO), 1 × N2 (Gibco), 2 mM Ultra-glutamine I (Lonza), 500 ng/ml R-Spondin-1(Peprotech), 100 ng/ml Noggin (Peprotech), 10 nM prostaglandin E2 (Cayman Chemicals), 3 μM SB202190 (Cayman Chemicals), 10 nM Gastrin I (Sigma-Aldrich), 50 ng/ml human recombinant EGF (Peprotech), 0.5 μM A83-01(Tocris), 10 mM Nicotinamide (Sigma-Aldrich), 1 mM N-acetylcysteine (Sigma-Aldrich), 1 × Pencillin/Streptomycin (Gibco) and 1 × Primocin (Invivogen), 10 μM Y-27632 supplemented with medium for the first 3 days. Normal colon organoid medium is identical to tumor organoid medium, except that it needs to be supplemented with 50% Wnt3a-conditioned medium. The Wnt3a-conditioned medium were obtained from L-Wnt3a cells. PDO was passaged approximately every 7–10 days using TrypLE Express (GIBCO) for about 10 min at 37 ℃ to dissociate the PDO into single cells. After dissociation, the cells were embedded in Matrigel (Corning, USA), and the medium was changed every 3–4 days.
TILs were generated as previous described [21]. Briefly, the tumor samples were minced into approximately 1–2 mm fragments after washed with ice-cold PBS buffer, and then put in a digestion medium consisting of 1 mg/ml collagenase (Sigma-Aldrich, USA) and 30U/ml DNase (Sigma, USA) at 37 °C for about 60 min. Following digestion, the cells were seeded into a 6-well plate containing T cell media supplemented with X-VIVO 15 (Lonza, USA), 2 mM Glutamax (Life Technologies, USA), 6000 U/mL IL-2 (Perprotech, USA), 10 ng/mL IL-15 (Perprotech, USA), 10 ng/mL IL-21 (Perprotech, USA), 1 × Primocin (Invivogen) and 30 ng/mL anti-human CD3 antibody (OKT-3, Miltenyi, Germany). TILs were cultured at 37 °C with 5% CO2 and passaged until enough TILs were available for screening tumor-specific T cells.
Peripheral blood lymphocytes
PBLs were obtained from freshly peripheral blood through Ficoll Paque plus (GE, DN25) density gradient centrifugation. Subsequently, the isolated PBLs were cryopreserved in liquid nitrogen for future utilization.
Histological analysis and immunohistochemistry staining of PDO
Organoids were collected (e.g., 6-well plate, collection of 1–2 well organoids, about 300–800 organoids) and subsequently fixed in 4% paraformaldehyde for 30 min at 4 °C, followed by gradient dehydration with ethanol and embedding in paraffin wax. The paraffin blocks were cut at a thickness of 5 μm using a paraffin microtome (Leica, RM2245) and subjected to routine hematoxylin–eosin (H&E) staining.
For immunohistochemistry staining, paraffin sections were routinely deparaffinized and incubated for 30 min in 3% H2O2 to inactivate endogenous peroxidase. Heat-induced antigen retrieval was performed using citric acid for 15 min at 95 °C. The sections were incubated with primary antibodies overnight at 4 °C. Then the HRP-conjugated second antibodies were stained for 1 h at room temperature (Supplementary Table S1). The bound antibody was visualized using the DAB detection kit (ZSGB-BIO, ZLI-9019) and counterstained with hematoxylin. The results were recorded using a light microscope (Olympus).
For immunofluorescent analysis, the organoids were fixed in 4% paraformaldehyde (DingGuo) for 15 min at 4 °C. Subsequently, they were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich). Then the organoids were incubated with primary antibodies overnight at 4 °C, followed by incubation with secondary antibodies for 1 h (Supplementary Table S1). Nuclear staining was performed using 4',6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma). The resulting images were captured using a confocal microscope (Zeiss, LSM780).
TILs and PBLs phenotypic characterization analysis
For CD39+CD103+ positive T cell analysis, TILs and PBLs from CRC patients were stained with anti-human CD3-FITC, anti-human CD8-PE, anti-human CD39-PE-Cy7, anti-human CD103-APC antibodies cocktails and 7AAD (Invitrogen). For anti-human CD8+PD1+ positive T cell analysis, TILs and PBLs were stained with anti-human CD3-FITC, anti-human CD8-PerCP5.5, anti-human CXCL13-PE antibodies cocktails and 7AAD. For CD8+PD1+ positive T cell analysis, TILs and PBLs were stained with anti-human CD3-FITC, anti-human CD8-PerCP5.5, anti-human PD-1-PE antibody cocktails and 7AAD (supplementary Table S2). Then samples were analyzed by gating of CD3+CD8+ T cells of all flow cytometry, the data were analyzed using FlowJo software.
Co-culture of tumor organoids and autologous TILs to generate tumor organoid-enriched T cells
Co-culture of single cell-derived organoids and autologous TILs were performed to enrich tumor-specific T cells. Briefly, two days before co-culture, tumor organoids embedded in Matrigel were stimulated with decitabine (DAC) cocktails (10 μM DAC (Sigma Aldrich), 100 U/mL IFNγ (ACRO) and 10 ng/mL TNF-α (ACRO)) for 48 h. Previous studies revealed that DAC cocktails can increase the expression of HLA by either suppressing DNA methylation or promoting the expression of HLA-associated TAP or LMP genes [22,23,24].
Next, the tumor organoids were dissociated into single cells using TrypLE Express and then resuspended in T cell medium. Prior to the experiment, a 96-well U-bottom plate was prepared by coating with 5 µg/mL anti-CD28 antibody in PBS (50 µL per well) and subsequently wrapping in parafilm overnight at 4 °C. The anti-CD28-coated plates were then washed twice with PBS. TILs were seeded at a concentration of 105 cells per well and co-cultured with autologous single cell organoids at a concentration of 104 cells per well at a 10:1 effector: target ratio. Co-cultures were performed in T cell medium (X-VIVO15 medium supplemented with 10% human AB serum, 2 mM glutamine, 150 U/ml IL-2, 10 ng/mL IL-15, 10 ng/mL IL-21, 1% penicillin–streptomycin, 20 μg/mL anti-PD-1 blocking antibody (Biolegend)) and cultured at 37 °C with 5% CO2. On the 7th day, co-cultures TILs were re-stimulated with fresh organoids at a ratio of 10:1 for an additional 7 days. The medium was half-changed every 1–2 days, and co-culture ended on day 14. TILs without tumor organoids stimulation were set up as a control.
Three-dimensional (3D) organoid killing assays
Organoids killing assay was performed in accordance to previously described with slight modification [25]. In brief, tumor organoids were stimulated with DAC cocktails for 48 h, and part of the organoids were dissociated into single cells and then counted to infer the number of tumor cells at a effector: target = 10:1 ratio. 1 mM of (Invitrogen) celltrace yellow cell proliferation kit (Invitrogen, cat. no. C34567) was used to stain organoids for 20 min. 3D tumor organoids and normal organoids were resuspended in T cell medium, and seeded in flat-bottom 96-well plate with 1 × 105 autologous oeT cells obtained by 14 days of tumor organoid co-culture. Green-fluorescent caspase 3/7 probe (Invitrogen, cat. no.C10723) was added during co-culture at 1:2000 dilution to visualize tumor cells undergoing apoptosis. After 20 h of co-culture, part of the cells (about 3 wells) were observed and photographed using fluorescence microscope, and part of the cells (about 6 wells) were used for flow detection of apoptotic cells. For MHC I/II blocking assay, tumor organoids were pretreated with 50 μg/mL pan-MHC I antibody (Biolegend, 311428) and pan-MHC II antibody (BD, 555556) blocking for 40 min before co-culture. Part of organoids dissociated into single cells with TrypLE Express, and the cells stained with anti-human CD3-FITC (Biolegend), anti-human CD8-PE (Biolegend), anti-human CD107a-PE-Cy7 (Biolegend) and 7AAD (supplementary Table S2). Then the samples were analyzed by flow cytometry.
Isolation of tumor-specific T cells after repeated stimulation with autologous tumor cells
To evaluate tumor reactivity of oeT cells, oeT cells were collected after 14 days of incubation with tumor organoids and resuspended in T cell medium. The oeT cells were then restimulated with tumor organoids and normal organoids at a 1:1 effector: target ratio, respectively. Subsequently, seeded in anti-CD28-coated plates in the presence of 20 mg/mL anti-PD-1 and co-cultured for 24 h. TILs without organoids stimulation and normal organoids stimulation were set up as controls. Cells were washed twice in PBS and stained with the following antibodies: anti-human CD3-FITC (Biolegend), anti-human CD8-PE (Biolegend), anti-human CD137-APC (Biolegend) and 7AAD for 30 min at 4 °C (Supplementary Table S2). Then the cells were washed twice with PBS. The expression of surface markers were quantified, and the CD8+CD137+ positive T cells were sorted into 96-well PCR plates using the Symphony S6 (BD) and analyzed using FlowJo software. Subsequently, the 96-well PCR plates were immediately stored at −80 °C.
T cell receptor sequencing and analysis
Single-cell RT-PCR is based on a previously described method [26]. The PCR primer sequences are listed in supplementary Table S3. Briefly, the first round of RT-PCR produces single-cell cDNA. The reaction conditions for the first-round RT-PCR were as follows: the reverse transcription reaction for 30 min at 50 °C; then 95 °C for 15 min and 30 cycles of 94℃ for 30 s, 52 °C for 30 s, 72 °C for 1 min; 72 °C for 10 min. Subsequently, RT-PCR cDNA from the first round was used as a template for the second round of TCRα/β PCR. The reaction conditions for the second round PCR are as follows: 98 °C for 1 min and 43 cycles of 98 °C for 10 s, 52 °C for 10 s, 72 °C for 45 s; 72 °C for 10 min. The PCR products were sequenced after gel cutting and purification. The TCR sequences were analyzed using the IMGT/V-Quest tool (http://www.imgt.org/).
TCR construction and transduction of PBLs
Construction of retroviral vectors encoding TCRs was used the methods as described previously [5, 27]. Briefly, we constructed the TCRs in a β chain-α chain order, fusing the TCRβ V-D-J of the Vβ regions to the mouse TCRβ constant chain and the TCRα V-J regions to the mouse TCRα-constant chain. As previously described, the mouse constant region was modified with a disulfide bond and a hydrophobic substitution [28,29,30]. Additionally, furin-SGSG-P2A sequence was introduced between the TCRβ chain and TCRα chain. The TCR sequences were synthesized after codon optimization (GenScript) and subsequently inserted into the lentivirus vector.
TCR transduction was performed in accordance to previously described with slight modification [22, 31]. In brief, human PBLs from healthy or autologous donors were stimulated with 50 ng/ml OKT3 in T cell media for 48 h before transduction. HEK293T cells were transfected with the lentivirus plasmid encoding TCRs and the envelope-encoding plasmid using CaCl2 to produce retroviral supernatants. After 48 h, the viral supernatants were harvested and then used to infect PBLs in the presence of polybrene (8 μg/mL). 3 days after infection, TCR-positive T cells were isolated through flow cytometry using a mouse TCR beta monoclonal antibody (Invitrogen, 17–5961-82). The sorted cells were subsequently confirmed using anti-human CD3-FITC (Biolegend, USA), anti-human CD8-PE (Biolegend, USA) and mouse TCR beta monoclonal antibody (eBioscience, USA).
Enzyme-linked immunospot assay (ELISPOT)
The human IFN-γ ELISPOT assay was performed according to the instructions provided by the manufacturer of the commercially available human IFN-γ ELISPOT kit (Dakewe, cat. No. 2110005, China). Briefly, before co-culturing, T cells rested in cytokine-free medium for 24 h and tumor organoids were treated with DAC cocktails for 48 h. A total of 1 × 104 T cells and 1 × 104 autologous organoids were incubated at 37 °C with 5% CO2 for 20 h in cytokine-free medium. Phytohemaggluinin (PHA) was added as a positive control, and medium was added as a negative control. The Immunospot plate reader (Cellular Technologies, USA) was used to analyze the number of IFN-γ positive spots.
Enzyme-linked immunosorbent (ELISA) assay
Human IFN-γ cytokine release was quantified using the human ELISA kit (Dakewe Biotech cat. no. 1110002, China). Briefly, T cells were rested in cytokine-free medium for 24 h, and organoids were stimulated with DAC cocktails for 48 h prior to co-culture. A total of 5 × 104 T cells and 5 × 104 autologous organoids were incubated together for 24 h in cytokine-free medium, and then the co-cultured supernatant was collected. IFN-γ cytokine secretion was measured using a commercially available human IFN-γ ELISA kit, following the manufacturer's protocols. In the MHC I blocking experiment, tumor organoids were pretreated with 50 μg/mL of pan-MHC I antibody (Biolegend, 311428) for 40 min before co-culture. The tumor recognition was tested in technical replicate experiments.
Statistics analysis
Data statistical analysis was performed using GraphPad Prism 7.0. The statistical analysis was conducted with a student's t-test. p-values < 0.05 were considered statistically significant, denoted by *(p < 0.05), **(p < 0.01), and ***(p < 0.001). All experiments were conducted more than three times independently.
Results
Tumor organoids retained the characterization of parent tumor tissue
Tumor samples were obtained from clinical patients with CRC. The clinical characteristics of the patients are outlined in Table 1. Freshly resected CRC tumor tissues were used to establish PDO as previously described [12]. Histological evaluation revealed significant morphological similarities between PDO and the tumor tissue that they were originally-derived. H&E staining illustrated that CRC tumor-derived organoids presented patient-specific heterogeneous morphologies ranging from hollow cystic structures to solid/compact structures (Fig. 1A, B). Multiple types of cells were found within the tumor organoid in a heterogeneous manner, such as paneth cells expressing Lysozyme (Lysz) and epithelial cells expressing Villin 1 (Vill), while a lower abundance of Mucin2 (Muc2)-expressing goblet cells were observed in the tumor organoids (Fig. 1C). An array of immunohistochemistry staining, including caudal type homeobox 2 (CDX2), cytokeratin 7 (CK7), cytokeratin 20 (CK20), Ki67 as well as antigen presenting molecule HLA-ABC, all showed similar staining patterns in the patient tumor tissue and matched PDO as shown in Fig. 1D. In conclusion, these data demonstrated that the PDO recapitulated the histological and genetic features of the original tumor tissues.
Tumor-specific T cells predominantly exist in TILs
Recent work has shown that tumor-specific T cells can be found within the CD39+CD103+ subset, PD-1+CD8+ subset and CXCL13+CD8+ subset of T cells from patients with epithelial cancers [32,33,34]. To evaluate the heterogeneity of tumor-specific T cells from CRC1 to CRC4 patients, we analyzed the frequency of CD39+CD103+, PD-1+CD8+ subset and CXCL13+CD8+ subset T cells in peripheral blood lymphocytes (PBLs) and matched tumor-infiltrating lymphocyte (TILs) fragments derived from CRC patients using flow cytometry.
Figure 2A, B show results from CD39+CD103+CD8+ T cells were found in the TILs and PBLs. The frequency of CD39+CD103+CD8+ T cells varied greatly among different CRC patients, with a higher frequency in TILs ranging from 24.9 to 43.1%. However, CD39+CD103+CD8+ T cells were present at a very low frequency in the PBLs, ranging from 0.15 to 7.7%. The frequency of PD-1+CD8+ T cells and CXCL13+ CD8+ T cells in TILs fragments ranged from 23.7 to 47.6% and 0.81 to 2.66% respectively. Although no statistically significant difference was observed between PBLs and TILs in these four CRC patients, there was a tendency for a lower frequency in PBLs (Fig. 2C–F). Collectively, these results indicate that tumor-specific T cells predominantly exist in TILs, which shows strong heterogeneity among different patients with CRC.
Organoid-TIL co-culture to generation of organoid-enriched T cells
In order to know whether tumor organoids may be used to enrich tumor-specific TILs. We generated PDO and TILs from randomly selected patients with CRC. Prior to co-cultured with autologous TILs, tumor organoids were pre-treatment with DAC cocktails to enhance antigen presentation. TILs were stimulated with autologous tumor organoids for two weeks to obtain oeT cells presented in Fig. 3A.We next assessed whether oeT cells have the ability to respond to autologous tumor PDO. Using an Enzyme-linked immunosorbent (ELISA) assay, oeT cells were stimulated with autologous tumor PDO and normal PDO, respectively. The ELISA outcome showed that oeT cells effectively induced interferon-γ (IFN-γ) secretion when stimulated with tumor PDO, but did not mount a cytotoxic response against normal organoids, demonstrating the tumor specificity of oeT cell-mediated cytotoxic response (Fig. 3B).
In order to determine whether tumor organoids can be effectively destroyed by oeT cells. We separately co-cultured tumor organoids and normal organoids with oeT cells for 20 h and quantified the number of apoptotic tumor cells by flow cytometry. To visualize the cytolytic activity of tumor-specific T cells, we labeled organoids with a tracer dye and imaged them in the presence of a green fluorescent apoptosis probe detecting active caspase-3/7 (‘‘caspase-3/7 probe’’). The results showed that tumor PDO induced more apoptosis than those from the other groups, whereas tumor organoids cultured without oeT cells or with oeT cells in the presence of blocking MHC class I and MHC class II antibodies group induced little apoptosis (Fig. 3C, D, Supplementary Figs. S1 and S2). Furthermore, the frequency of T cell effector markers CD107a was higher in the tumor PDO group (CD107a: 9%) than those from the normal PDO and PDO alone group (CD107a: 4.4%, 1.8%: respectively) (Fig. 3E, F, Supplementary Fig. S3). Taken together, these data demonstrated that oeT cells had better anti-tumor ability.
Tumor-specific T cells were isolated from co-cultured oeT cells using CD137 expression
In order to obtain tumor-specific T cells and their TCRs, oeT cells were stimulated with tumor PDO again, normal PDO and TILs alone as negative controls. T cell responses were assessed through the upregulation of CD137 and secretion of IFN-γ using an enzyme-linked immunospot (ELISPOT) assay. We found that oeT cells co-cultured with tumor PDO exhibited a significantly higher quantity of IFN-γ positive cell spots compared to normal PDO or TIL alone in CRC patients (Fig. 4A, B).
Single CD8+ CD137+ T cells were isolated and their TCR sequences were amplified using single-cell PCR in 96-well plates.
The tumor-specific TCR may be the most dominant TCR on CD8+ CD137+ oeT cells. In order to eliminate the possibility of non-specific TCR amplification, oeT cells stimulated with normal PDO and TIL alone were used as a negative control. Meanwhile,
the flow cytometry results showed a significantly higher frequency of tumor-specific T cells in the tumor PDO group compared to both the TIL alone and normal PDO groups among CRC patients. However, little difference in the frequency of tumor-specific T cells between the TIL alone group and normal PDO group (Fig. 4C, D).
Construction and functional validation of tumor-specific TCR-Ts
To determine if oeT cells are clonally proliferated subsets of tumor-specific T cells, we isolated CD8+ CD137+ T cells from both the TIL alone group and the oeT cells from the tumor PDO-stimulated group. The TCR α/β chain pairs were then identified using single-cell RT-TCR sequencing. The results showed that no TCR was represented more than 4.3% of the TIL alone group in CRC patients, demonstrating the polyclonal biological characteristics of the subpopulations (Fig. 5A). However, the TCRs of tumor PDO group showed significant oligoclonal. In CRC1, oeT cells were predominantly composed of TCR1 and TCR2, accounting for 25.6% and 17.8% of the T cell populations, respectively. In contrast, the TIL alone group had a much lower proportion of TCR1 and TCR2, representing only 4.3% and 3.6% of these subsets. In CRC2, the frequency of TCR9 was 1.1% in TIL alone and increased to 8.2% in oeT cells, indicating a more than 7-folds clonal expansion after incubation with autologous tumor PDO in vitro. In CRC3, the TCR11 was undetectable among the top five in TIL alone, but it was enriched to 11.4% in oeT cells, demonstrating a more than 28-folds clonal expansion (Table 2). In CRC4, TCR17 underwent more than a 5-folds clonal proliferation compared to before stimulation, increasing from 2.4% in TIL alone to 13.5% in oeT cells after co-culture with autologous tumors (Fig. 5A).
We sequenced the Vβ and Vα chains of TCRs in oeT cells and chose the top five to create chimeric TCRs (Table 2). For enhanced expression of introduced TCRs in T cells, the TCRs were designed in a β/α chain sequence and the constant regions were replaced with mouse-derived counterparts that had been altered to include interchain disulfide bonds and hydrophobic modifications, as previously documented [35]. The TCRs were lentivirus-transduced into healthy donor T cells to generate TCR-Ts. Then these T cells were sorted and achieved a positive rate of more than 76.25%, as shown in Fig. 5B. We then investigated whether these TCR-Ts could specifically identify and mediate effector functions in response to autologous tumor PDO through releasing cytokines. Engineered TCR1-T cells and TCR3-T cells exhibited increased levels of IFN-γ secretion and specific cytotoxicity toward autologous tumor PDO compared to normal PDO (Fig. 5C, D). In CRC2, three out of the five selected TCRs could specifically recognize autologous tumor cells, as evidenced by the IFN-γ production of these engineered TCR-T cells when incubated with autologous tumor cells (Fig. 5E, supplementary Fig. S4). Co-culture of the engineered TCR12-Ts and TCR13-Ts cells with autologous tumor organoids induced significant IFN-γ secretion. While incubating them with normal organoids did not induce detectable IFN-γ secretion. Additionally, we revealed that that TCR14-Ts displayed significantly inhibited IFN-γ secretion levels with MHC class I blocking antibody, indicating that the tumor recognition by TCRs is highly MHC-dependent (Fig. 5F). These findings provided direct evidence that the ability of the TCRs identified in oeT cells were sufficient to confer the tumor-targeting ability to engineer TCR-T cells.
Discussion
Although previous studies have documented the sustained regression of metastatic colorectal cancer and metastatic melanoma through the adoptive transfer of in vitro expanded autologous TILs, the efficacy of this therapy is limited in numerous solid tumors due to the presence of a large number of bystander T cells and exhausted cell phenotypes [36,37,38]. Therefore, efficiently identifying tumor-specific TCRs and introduced into peripheral T cells to generate tumor-specific TCR-Ts is an optimal strategy for clinical therapy. In this study, we developed a co-culturing platform for the enrichment and expansion of tumor-specific TILs in vitro through repeated stimulation of autologous tumor organoids from CRC patients. The identification of tumor-specific TCRs by flow sorting of CD8+CD137+ T cells and the TCRs subsequent introduction of PBLs showed specific tumor reactivity and cytotoxicity. The organoid-autologous TILs co-culture is an effective system for identifying tumor-specific TCRs for TCR-T therapy. Further experiments will need to assess the effectiveness of tumor-specific TCRs for adoptive cell therapy.
Previous studies have reported the generation of tumor-specific T cells by co-incubating tumor organoids with peripheral blood [39]. However, this method can be overall low-yield, possibly due to the low frequency of tumor-specific T cells in peripheral blood. In this study, we systematically analyzed the frequency of tumor-specific T cells in peripheral blood and matched TILs from the same patient. we found that the frequency of CD39+CD103+ T cells in the TILs was much higher than in the matched peripheral blood from the same patient, which suggests that TILs have a higher frequency of tumor-specific T cells compared to those found in peripheral blood. The consistent results with previous reports, neoantigen-specific T cells that are occurring multiple times in TILs and found at a frequency greater than twice the limit of detection in the peripheral blood [20, 32]. Hence, the use of TILs, which greatly enrich tumor-specific T cells, can improve the efficiency of identifying tumor-specific TCRs through organoid stimulation.
Previous studies have focused on screening tumor-specific TCRs using predicted tumor antigens [5, 40]. However, there is less evidence about screening tumor-specific TCRs using natural antigens from tumor organoids. Therefore, in terms of whether tumor-specific TCRs could be enriched by tumor organoids-autologous TILs immune co-cultures, we found that the proportion of CD137-positive T cells after tumor organoids co-cultures was significantly higher than the control group. Meanwhile, we demonstrated a high degree of clonal expansion of TCRs from CD8+ oeT after sequencing analysis. These results showed that tumor-specific TCRs can be obtained by co-culture of autologous tumor organoids with TILs in vitro. And their TCRs can adequately recognized autologous tumor organoids in a MHC-dependent manner. Previous studies have demonstrated that tumor organoids can accurately reflect the mutanome of their parental tumor tissue, as well as the diversity in MHC class I peptide presentation, including tumor antigens [16, 41]. Combined with these results, we showed that tumor organoids can be perform as natural tumor antigens for enrichment of tumor-specific TCRs.
There are some limitations in this study. Firstly, previous study showed that tumor organoids express MHC class II proteins in pancreatic cancer, suggesting that the tumor cells present MHC class II antigens for CD4+ T cells [42]. Our study mainly focuses on tumor-specific CD8+ T cells and their TCRs specific response and cytotoxicity to autologous tumor cells after organoid enrichment. Therefore, we plan to investigate whether colorectal organoids co-cultured with TILs can effectively enrich CD4+ tumor-specific T cells in the future. In addition, we identified tumor-specific TCRs using low-throughput single-cell RT-PCR and Sanger sequencing. This approach is limited in comprehending the comprehensive information of TCR repertoire after stimulation with organoid. High-throughput single-cell transcriptome sequencing will be incorporated into future studies, which would provide us with a more complete picture of the genetic composition of tumor-specific TCRs. Finally, in this study using tumor organoids as the source of natural tumor antigen to screen tumor-specific TCRs, however, the specific antigen epitope recognized by tumor-specific TCRs remains unknown. A variety of approaches have been used to address this issue, including T cell trogocytosis and baculoviral peptide libraries [43, 44]. These methods provide valuable insights for the identification of specific antigen epitopes after known cognate tumor-specific TCRs.
Data availability statement
The data generated in this study are available within the article and its supplementary data files.
References
Parkhurst MR, Yang JC, Langan RC et al (2011) T cells targeting carcinoembryonic antigen can mediate regression of metastatic colorectal cancer but induce severe transient colitis. Mol Ther 19:620–626. https://doi.org/10.1038/mt.2010.272
Robbins PF, Kassim SH, Tran TL et al (2015) A pilot trial using lymphocytes genetically engineered with an NY-ESO-1-reactive T-cell receptor: long-term follow-up and correlates with response. Clin Cancer Res 21:1019–1027. https://doi.org/10.1158/1078-0432.CCR-14-2708
Meng F, Zhao J, Tan AT et al (2021) Immunotherapy of HBV-related advanced hepatocellular carcinoma with short-term HBV-specific TCR expressed T cells: results of dose escalation, phase I trial. Hep Int 15:1402–1412. https://doi.org/10.1007/s12072-021-10250-2
Leidner R, Sanjuan Silva N, Huang H et al (2022) Neoantigen T-cell receptor gene therapy in pancreatic cancer. N Engl J Med 386:2112–2119. https://doi.org/10.1056/NEJMoa2119662
Parkhurst M, Gros A, Pasetto A, Prickett T, Crystal JS, Robbins P, Rosenberg SA (2017) Isolation of T-cell receptors specifically reactive with mutated tumor-associated antigens from tumor-infiltrating lymphocytes based on CD137 expression. Clin Cancer Res 23:2491–2505. https://doi.org/10.1158/1078-0432.CCR-16-2680
Braunlein E, Lupoli G, Fuchsl F et al (2021) Functional analysis of peripheral and intratumoral neoantigen-specific TCRs identified in a patient with melanoma. J Immunother Cancer 9. https://doi.org/10.1136/jitc-2021-002754
Purbhoo MA, Li Y, Sutton DH et al (2007) The HLA A*0201-restricted hTERT(540–548) peptide is not detected on tumor cells by a CTL clone or a high-affinity T-cell receptor. Mol Cancer Ther 6:2081–2091. https://doi.org/10.1158/1535-7163.MCT-07-0092
Willimsky G, Beier C, Immisch L, Papafotiou G, Scheuplein V, Goede A, Holzhutter HG, Blankenstein T, Kloetzel PM (2021) In vitro proteasome processing of neo-splicetopes does not predict their presentation in vivo. eLife 10. https://doi.org/10.7554/eLife.62019
Tran E, Robbins PF, Rosenberg SA (2017) “Final common pathway” of human cancer immunotherapy: targeting random somatic mutations. Nat Immunol 18:255–262. https://doi.org/10.1038/ni.3682
Laumont CM, Vincent K, Hesnard L et al. (2018) Noncoding regions are the main source of targetable tumor-specific antigens. Sci Translat Med. 10. https://doi.org/10.1126/scitranslmed.aau5516
Wang W, Yuan T, Ma L et al (2022) Hepatobiliary tumor organoids reveal HLA Class I neoantigen landscape and antitumoral activity of neoantigen peptide enhanced with immune checkpoint inhibitors. Adv Sci 9:e2105810. https://doi.org/10.1002/advs.202105810
Cattaneo CM, Dijkstra KK, Fanchi LF, Kelderman S, Kaing S, van Rooij N, van den Brink S, Schumacher TN, Voest EE (2020) Tumor organoid-T-cell coculture systems. Nat Protoc 15:15–39. https://doi.org/10.1038/s41596-019-0232-9
Beshiri ML, Tice CM, Tran C et al (2018) A PDX/organoid biobank of advanced prostate cancers captures genomic and phenotypic heterogeneity for disease modeling and therapeutic screening. Clin Cancer Res 24:4332–4345. https://doi.org/10.1158/1078-0432.Ccr-18-0409
Broutier L, Mastrogiovanni G, Verstegen MM et al (2017) Human primary liver cancer-derived organoid cultures for disease modeling and drug screening. Nat Med 23:1424–1435. https://doi.org/10.1038/nm.4438
Kopper O, de Witte CJ, Lõhmussaar K et al (2019) An organoid platform for ovarian cancer captures intra- and interpatient heterogeneity. Nat Med 25:838–849. https://doi.org/10.1038/s41591-019-0422-6
Demmers LC, Kretzschmar K, Van Hoeck A et al (2020) Single-cell derived tumor organoids display diversity in HLA class I peptide presentation. Nat Commun 11:5338. https://doi.org/10.1038/s41467-020-19142-9
Rosenberg SA, Restifo NP (2015) Adoptive cell transfer as personalized immunotherapy for human cancer. Science (New York, NY) 348:62–68. https://doi.org/10.1126/science.aaa4967
Stevanović S, Draper LM, Langhan MM et al (2015) Complete regression of metastatic cervical cancer after treatment with human papillomavirus-targeted tumor-infiltrating T cells. J Clin Oncol 33:1543–1550. https://doi.org/10.1200/jco.2014.58.9093
Rosenberg SA, Packard BS, Aebersold PM et al (1988) Use of tumor-infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic melanoma: a preliminary report. N Engl J Med 319:1676–1680. https://doi.org/10.1056/nejm198812223192527
Lowery FJ, Krishna S, Yossef R et al (2022) Molecular signatures of antitumor neoantigen-reactive T cells from metastatic human cancers. Science 375:877–884. https://doi.org/10.1126/science.abl5447
Topalian SL, Muul LM, Solomon D, Rosenberg SA (1987) Expansion of human tumor infiltrating lymphocytes for use in immunotherapy trials. J Immunol Methods 102:127–141. https://doi.org/10.1016/s0022-1759(87)80018-2
Tan Q, Zhang C, Yang W, Liu Y, Heyilimu P, Feng D, Xing L, Ke Y, Lu Z (2019) Isolation of T cell receptor specifically reactive with autologous tumour cells from tumour-infiltrating lymphocytes and construction of T cell receptor engineered T cells for esophageal squamous cell carcinoma. J Immunother Cancer 7:232. https://doi.org/10.1186/s40425-019-0709-7
Serrano A, Tanzarella S, Lionello I, Mendez R, Traversari C, Ruiz-Cabello F, Garrido F (2001) Rexpression of HLA class I antigens and restoration of antigen-specific CTL response in melanoma cells following 5-aza-2’-deoxycytidine treatment. Int J Cancer 94:243–251. https://doi.org/10.1002/ijc.1452
Seliger B, Harders C, Lohmann S, Momburg F, Urlinger S, Tampe R, Huber C (1998) Down-regulation of the MHC class I antigen-processing machinery after oncogenic transformation of murine fibroblasts. Eur J Immunol 28:122–133. https://doi.org/10.1002/(SICI)1521-4141(199801)28:01%3c122::AID-IMMU122%3e3.0.CO;2-F
Liu T, Tan J, Wu M et al (2021) High-affinity neoantigens correlate with better prognosis and trigger potent antihepatocellular carcinoma (HCC) activity by activating CD39(+)CD8(+) T cells. Gut 70:1965–1977. https://doi.org/10.1136/gutjnl-2020-322196
Dash P, Wang GC, Thomas PG (2015) Single-cell analysis of T-cell receptor alphabeta repertoire. Methods Mol Biol 1343:181–197. https://doi.org/10.1007/978-1-4939-2963-4_15
Tran E, Ahmadzadeh M, Lu YC et al (2015) Immunogenicity of somatic mutations in human gastrointestinal cancers. Science 350:1387–1390. https://doi.org/10.1126/science.aad1253
Cohen CJ, Zhao Y, Zheng Z, Rosenberg SA, Morgan RA (2006) Enhanced antitumor activity of murine-human hybrid T-cell receptor (TCR) in human lymphocytes is associated with improved pairing and TCR/CD3 stability. Can Res 66:8878–8886. https://doi.org/10.1158/0008-5472.CAN-06-1450
Cohen CJ, Li YF, El-Gamil M, Robbins PF, Rosenberg SA, Morgan RA (2007) Enhanced antitumor activity of T cells engineered to express T-cell receptors with a second disulfide bond. Can Res 67:3898–3903. https://doi.org/10.1158/0008-5472.CAN-06-3986
Lu YC, Zheng Z, Robbins PF et al (2018) An efficient single-cell RNA-Seq approach to identify neoantigen-specific T Cell receptors. Mol Ther 26:379–389. https://doi.org/10.1016/j.ymthe.2017.10.018
Yossef R, Tran E, Deniger DC et al. (2018) Enhanced detection of neoantigen-reactive T cells targeting unique and shared oncogenes for personalized cancer immunotherapy. JCI Insight. 3. https://doi.org/10.1172/jci.insight.122467
Duhen T, Duhen R, Montler R et al (2018) Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. Nat Commun 9:2724. https://doi.org/10.1038/s41467-018-05072-0
Inozume T, Hanada K, Wang QJ, Ahmadzadeh M, Wunderlich JR, Rosenberg SA, Yang JC (2010) Selection of CD8+PD-1+ lymphocytes in fresh human melanomas enriches for tumor-reactive T cells. J Immunother 33:956–964. https://doi.org/10.1097/CJI.0b013e3181fad2b0
Zheng C, Fass JN, Shih YP et al (2022) Transcriptomic profiles of neoantigen-reactive T cells in human gastrointestinal cancers. Cancer Cell 40:410–423. https://doi.org/10.1016/j.ccell.2022.03.005
Tsuji T, Yoneda A, Matsuzaki J, Miliotto A, Ryan C, Koya RC, Odunsi K (2018) Rapid construction of antitumor T-cell receptor vectors from frozen tumors for engineered T-cell therapy. Cancer Immunol Res 6:594–604. https://doi.org/10.1158/2326-6066.CIR-17-0434
Tran E, Robbins PF, Lu YC et al (2016) T-cell transfer therapy targeting mutant KRAS in cancer. N Engl J Med 375:2255–2262. https://doi.org/10.1056/NEJMoa1609279
Rosenberg SA, Yang JC, Sherry RM et al (2011) Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res 17:4550–4557. https://doi.org/10.1158/1078-0432.CCR-11-0116
Goff SL, Dudley ME, Citrin DE et al (2016) Randomized, prospective evaluation comparing intensity of lymphodepletion before adoptive transfer of tumor-infiltrating lymphocytes for patients with metastatic melanoma. J Clin Oncol 34:2389–2397. https://doi.org/10.1200/JCO.2016.66.7220
Dijkstra KK, Cattaneo CM, Weeber F et al (2018) Generation of Tumor-Reactive T Cells by Co-culture of Peripheral Blood Lymphocytes and Tumor Organoids. Cell 174:1586–1598. https://doi.org/10.1016/j.cell.2018.07.009
Malekzadeh P, Yossef R, Cafri G et al (2020) Antigen experienced T cells from peripheral blood recognize p53 neoantigens. Clin Cancer Res 26:1267–1276. https://doi.org/10.1158/1078-0432.CCR-19-1874
Parikh AY, Masi R, Gasmi B et al (2023) Using patient-derived tumor organoids from common epithelial cancers to analyze personalized T-cell responses to neoantigens. Cancer Immunol Immunother: CII 72:3149–3162. https://doi.org/10.1007/s00262-023-03476-6
Meng Q, Xie S, Gray GK et al (2021) Empirical identification and validation of tumor-targeting T cell receptors from circulation using autologous pancreatic tumor organoids. J Immunother Cancer. 9. https://doi.org/10.1136/jitc-2021-003213
Joglekar AV, Li G (2021) T cell antigen discovery. Nat Methods 18:873–880. https://doi.org/10.1038/s41592-020-0867-z
Li G, Bethune MT, Wong S et al (2019) T cell antigen discovery via trogocytosis. Nat Methods 16:183–190. https://doi.org/10.1038/s41592-018-0305-7
Acknowledgements
We thank Dr. Zhaoya Gao for his assistance of providing surgical samples and clinical characterizations of the patient. We thank Dr. Tianxing Liu for critical input on the manuscript revision. Thanks Hongkui Deng and Yun Bai for their helpful and critical suggestions during the course of the study.
Funding
This work was supported by the National Key Research and Development Program of China Grant (2022YFA1103103). and the National Natural Science Foundation of China (32071413 and 82373261).
Author information
Authors and Affiliations
Contributions
ZLL, LSM and ZYG designed the study, ZLL, LSM and XYW collected, analyzed, and interpreted data, ZLL, LSM and TXL wrote the manuscript. PZZ, QJZ, YXL, YS and XC analyzed data and provided statistical interpretation. All authors read and approved the final manuscript.
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Ethics approval and consent to participate
This study was approved by the Medical Ethical Committee of the Peking University Shougang Hospital and written informed consent was obtained from all patients.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Li, Z., Ma, L., Gao, Z. et al. Identification and validation of tumor-specific T cell receptors from tumor infiltrating lymphocytes using tumor organoid co-cultures. Cancer Immunol Immunother 73, 164 (2024). https://doi.org/10.1007/s00262-024-03749-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00262-024-03749-8