Development of the T-ALLiPSC-based therapeutic cancer vaccines for T-cell acute lymphoblastic leukemia

The T-cell acute lymphoblastic leukemia (T-ALL) is a kind of hematological malignancy in children. Despite the significant improvement in the cure rate of T-ALL upon treatment with chemotherapy regimens, steroids, and allotransplantation there are relapses. This study focuses on the tumor-specific therapeutic vaccines derived from the induced pluripotent stem cells (iPSC) to address the issue of T-ALL recurrence. Patient-derived tumor cells and healthy donor cells were reprogrammed into the iPSCs and the RNA-seq data of the T-ALL-iPSCs and H-iPSCs were analyzed. In vitro, the whole-cell lysate antigens of iPSCs were prepared to induce the dendritic cells (DC) maturation, which in turn stimulated the tumor-specific T cells to kill the T-ALL tumor cells (Jurkat, CCRF-CEM, MOLT-4). The cytotoxic T lymphocyte stimulated by the DC-loaded T-ALL-iPSC-derived antigens showed specific cytotoxicity against the T-ALL cells in vitro. In conclusion, the T-ALL-iPSC-based therapeutic cancer vaccine can elicit a specific anti-tumor effect on T-ALL.


Introduction
Childhood acute lymphoblastic leukemia (ALL) is a genetically heterogeneous cancer accounting for 10-15% of the T-cell ALL (T-ALL) cases. The T-ALL event-free survival rate (EFS) is as high as 80-90% [1]. It has a higher risk of relapse than the B-lineage ALL with has a poor outcome after relapse [2]. Here, the potential of using the tumor-associated proteins in the induced pluripotent stem cells (iPSC) was explored as the basis of a T-ALL therapeutic vaccine.
A therapeutic cancer vaccine is based on the principle of inducing a specific anti-tumor immune response by introducing a tumor-related antigen. Generally, the antigens of the cancer vaccines constitute the tumor cells, tumor antigen-related proteins and peptides synthesized in vitro, and antigen-related genes [3,4]. Studies performed a century ago showed that immunizing mice with embryonic tissues triggered the rejection of the transplanted tumors [5]. Furthermore, the embryonic stem cells (ESC)-vaccinated mice received obvious anti-tumor immunity, protecting them from lung cancer formation and development [6]. Emerging data have demonstrated ESCs to share common transcriptome features and antigens with the tumor cells; however, ethical concerns limit their usage in the cancer vaccines.
Since the discovery of iPSCs [7,8], the autologous iPSCs have been found to share almost the same gene expression and surface markers as ESCs [9,10]. In addition, the tumorigenicity and immunogenicity of the autologous iPSCs indicate their potential efficacy in cancer vaccination [11,12]. This highlights that iPSCs can provide more accurate and representative patients' tumor immunogen than the nonautologous ESCs [12].
Latest studies have shown that [11,12] iPSCs inactivation can be harnessed as a cancer vaccine inducing extensive anti-tumor immune responses. Through RNA sequencing (RNA-seq), the researchers found that iPSCs from both humans and mice express tumor-associated antigens, indicating that all of these iPSCs could be utilized as an anticancer vaccine. Injecting the iPSC-based vaccine can effectively prevent the growth of breast cancer, mesothelioma, and melanoma in tumor-bearing mice. The IPSC-based vaccine injection can also effectively kill the tumor tissue in vivo. These results indicated that the iPSC-based vaccine can induce an effective immune response in the mice and exert widespread killing effects on the different types of tumors. Compared to the current immunotherapy strategy, the iPSC vaccine can reactivate the immune system and target cancer without any side effects, proving a novel method for developing a clinical cancer vaccine.
This study first reprogrammed the tumor cells from the T-ALL patients into iPSCs, preparing the whole-cell lysate antigens of iPSCs for obtaining a specific T-ALL-iPSC-based vaccine. Subsequently, it was presented to the T cells through the dendritic cells (DC), stimulating specific immune reactions against T-ALL.

Cell culture
The human T-ALL cell lines (Jurkat, CCRF-CEM, MOLT-4) and NSCLC cell line NCI-H1299 were cultured in the RPMI 1640 medium supplemented with 10% FBS. The human cervix adenocarcinoma, HeLa cells were cultured in Eagle's minimum essential medium supplemented with 10% FBS and 1% NEAA (10 mM). The breast cancer cell line, MCF-7 cells were cultured in the DMEM supplemented with 10% FBS.

T-ALL reprogramming and preparation of the T-ALL-iPSC lysates
The peripheral blood mononuclear cells (PBMC) from the T-ALL patients and healthy donors were infected with the CytoTune-iPS 2.0 Sendai virus reprogramming Kit (Thermo Fisher). The obtained ESC-like clones were identified for pluripotency, using methods like alkaline phosphatase staining [13], flow cytometry (2.4), and immunofluorescence staining [13] for the pluripotency markers and teratoma formation [14]. The T-ALL-iPSCs were cultured with mTeSR plus medium (STEMCELL Technologies).
To prepare cell lysates, the iPSCs derived from the healthy cells (H-iPSCs) (1 × 107/ml) and T-ALL-iPSCs (1 × 107/ml) were frozen in liquid nitrogen and disrupted by three freeze-thaw cycles. The cell lysates were then spun at 13,000×g for 20 min, and the supernatants were collected as iPSC antigens. The total protein concentration of the cell lysates was quantified using the BCA Protein Assay Kit (Beyotime Institute of Biotechnology, China).

Generation of the different types of DC-CTLs
The DCs were prepared from the monocytes of the healthy donors. The PBMC were isolated by Lymphoprep (Stemcell, USA) density gradient centrifugation at 800×g for 15 min at 20 ℃. After 2 h of incubation, the adherent cells were cultured in an Alys-505 complete culture medium (Zhuhai Baso Cell Science & Technology Co. Ltd, China) supplemented with 100 ng/ml GM-CSF and 30 ng/ml IL-4 induction of DCs. The non-adherent cells were cultured in the Alys-505 complete culture medium containing 1000 U/ml IL-2 and 200 ng/ml CD3 monoclonal antibody for generating the CTLs.
On day 3, the immature DCs were treated with 30 µg/ml of different antigens and 1000 U/ml TNF-α. On day 7, the CTLs were co-cultured with autologous DCs at a 5:1 ratio. There were four groups, T group (only CTLs), DC-T group (CTLs were co-cultured with DCs), DC H-iPSC -T group (CTLs were co-cultured with H-iPSC lysates-stimulated DCs), and DC T-ALL-iPSC -T group (CTLs were co-cultured with T-ALL-iPSC lysates-stimulated DCs). All the groups were cultured at 37˚C in a humidified 5%CO2 incubator and passaged every 3 days in an Alys-505 complete culture medium with 1000 U/ml IL-2.

Flow cytometry
The T-ALL-iPSC were stained with PE anti-human TRA-1-60 and APC anti-human SSEA-4 (Biolegend, USA) for 20 min at 4 °C and analyzed by the flow cytometry software. T cells were stained with FITC anti-human CD3, PerCp/ Cyanine5.5 anti-human CD4, APC anti-human CD8a, PE anti-human CD25, PE anti-human PD-1, and PE anti-human CD57 (Biolegend, USA) for 20 min at 4 °C and analyzed by the flow cytometry software.

In vitro cytotoxicity assay
The CTLs from each group were examined for the cytotoxic activity toward the target tumor cells by performing a lactate dehydrogenase (LDH) release assay using the LDH cytotoxicity assay kit (Beyotime Institute of Biotechnology). The CTLs and tumor cells were co-cultured at the effector-to-target (E/T) ratios of 5:1, 10:1, and 20:1 for 24 h. The specific lysis is calculated as follows: Percentage of cytotoxicity = [ (experimental release-effector spontaneous release-target spontaneous release)]/[(target maximum release-target spontaneous release)] × 100%.

Cytokine release assay
The T cells were co-cultured with or without DCs for 24 h. The supernatants from each group were collected for assessing IL-6, IL-13, IFN-γ, and TNF-α (Biolegend Human Th Cytokine Panel (12-plex) with V-bottom Plate V02, USA) using flow cytometry.
CTLs from each group were co-incubated with the target cells at an E:T ratio of 10:1. Following 24 h of incubation at 37 °C, the supernatant was harvested and the concentration of the released IFN-γ was measured using the human IFNγ-Precoated ELISA Kit (Dakewe Bio-engineering Co. Ltd, Shenzhen, China).

Degranulation assay
The CTLs from each group were co-cultured with the target cells at an E:T ratio of 10:1. The PE anti-CD107a antibody (Biolegend, USA) was included during the culture. After 1 h, 2 μM monensin (Biolegend, USA) was added to the co-cultured system and incubated for an additional 4 h and then the cells were stained with the APC anti-human CD8a and analyzed using flow cytometry. All the CD8 + CD107a + cells were regarded as the degranulated T cells.

Analysis of RNA-seq data
The human RNA-seq data for the cancer cell lines were downloaded from the Cell Model Passports Database (https:// cellm odelp asspo rts. sanger. ac. uk/). The cancerrelated genes list was downloaded from the Network of Cancer Genes (http:// ncg. kcl. ac. uk/ cancer_ genes. php). The list of over-expressed genes in T-ALL was screened from the Oncomine's Haferlach Leukemia Dataset [15]. The heatmap was generated using GENE-E.

Statistical analysis
The GraphPad Prism 5 software was used for data analysis. All data are presented as the mean ± the standard deviations. The one-way ANOVA with Bonferroni's post hoc test was used to analyze the differences between the groups. p < 0.05 indicates a statistical difference.

The identification of pluripotency of the T-ALL-iPSC
The pluripotency of the T-ALL-iPSC was identified by the alkaline phosphatase staining (Fig. 1A) and the expression of pluripotent markers (Fig. 1B-D). Figure 1E shows the karyotype analysis of the T-ALL-iPSC. When the T-ALL-iPSC were transplanted to the immunodeficient mice, these cells could successfully form the teratomas with three germ layers (Fig. 1F). Collectively, these data show that the patient's T-ALL cells could be successfully reprogrammed to T-ALL-iPSCs.

The T-ALL-iPSCs have more similar T-ALL-related gene expression profiles than the H-iPSCs
Human-induced pluripotent stem cells (hiPSCs) have epigenetic memory and biased propensities for differentiation [16]. To investigate whether the autologous iPSCs have the potential to yield tumor-specific therapeutic vaccines, the transcriptome expression correlation matrix was used to analyze the similarity between the T-ALL-iPSC, H-iPSC, and T-ALL cell lines (Jurkat, CCRF-CEM, MOLT-4).
The T-ALL-iPSC had a higher similarity with T-ALL cell lines than the H-iPSC (H-iPSC-1, H-iPSC-2) ( Fig. 2A). The upregulation of a subset of genes was identified in the T-ALL-iPSCs which were over-expressed in the T-ALL cell lines (Fig. 2B). Collectively, these data show that the T-ALL-iPSCs were more similar to the T-ALL-related gene profiles than the H-iPSC, suggesting that the T-ALL-iPSCs are more suitable for yielding cancer vaccines for T-ALL than the H-iPSCs.

DC-CTL phenotype and cytokine secretion
The DCs are currently known as the strongest professional antigen-presenting cells activating the resting T cells [17]. The immature DCs were loaded with the cell lysates of iPSCs, and the ability of the ex vivo-generated mature DCs to stimulate the T-cell function was evaluated (Fig. 3). The proportion of the activated T cells (Fig. 3A), exhausted T cells (Fig. 3B), and senescent T cells (Fig. 3C) were analyzed in the different groups. As shown in Fig. 3A, the proportion of the CD3 + CD25 + T cells and CD3 + CD8 + CD25 + T cells increased in the DC-T group, H-iPSC group, and T-ALL-iPSC group compared to the T group.
We measured the concentrations of the cytokines IL-6 ( Fig. 3D), IL-13 (Fig. 3E), IFN-γ (Fig. 3F), and TNF-α (Fig. 3G) in the supernatants secreted by the CTLs. The CTLs stimulated by the DCs had a stronger function and cytokine secretion than the T group, whereas there were no significant differences between the DC H-iPSC -T and DC T-ALL-iPSC -T groups.

The DCT -ALL-iPSC -Ts and DCH -iPSC -Ts had similar cytotoxicity against the non-T-ALL tumor cells
The iPSCs were reported to share the gene expression profiles with the cancer cells [11,12]. We first identified cytotoxic activity against non-T-ALL cells. The cytotoxicity of CTLs against the non-T-ALL cells including Hela (Fig. 4A), MCF-7 (Fig. 4B), and NCI-H1299 (Fig. 4C) was examined and the level of IFN-γ secreted by CTLs was determined by ELISA at an E:T ratio of 10:1 (Fig. 4D, E). The percentage of CD8 + CD107a + (degranulation assay) was measured by flow cytometry at an E:T ratio of 10:1 (Fig. 4G-I). As shown in Fig. 4, the DC-stimulated CTL groups had higher cytotoxic activity than the non-DC-stimulated CTL group, but no significant differences were found between the DC H-iPSC -T and the DC T-ALL-iPSC -T groups.

The DCT -ALL-iPSC -Ts exhibited specific cytotoxicity against the T-ALL cells in vitro
To assess whether the T-ALL-iPSCs were effective vaccines against the T-ALL cells and whether the stimulated T cells contained antigen-specific CTLs, the cytotoxic activity of the DC T-ALL-iPSC -CTLs against T-ALL cells were analyzed in vitro. The cytotoxicity of the DC T-ALL-iPSC -T group against the T-ALL cells including Jurkat (Fig. 5A), CCRF-CEM (Fig. 5B), and MOLT-4 ( Fig. 5C) was significantly higher compared to the DC H-iPSC -T group. To further verify the cytotoxicity, the level of IFN-γ secreted by the CTLs co-cultured with the T-ALL cells at an E:T ratio of 10:1 was measured (Fig. 5D, E). Finally, the percentage of the CD8 + CD107a + cells was detected when the CTL killed the T-ALL tumor cells (Fig. 5G-I). Collectively, these data show that the DC T-ALL-iPSC -Ts exhibited specific cytotoxicity against the T-ALL cells.

Discussion
Both tumorigeneses, as well as tumor development, comprise highly proliferative and low-immune cells that escape the surveillance of the immune system. Therefore, targeting tumors by reactivating the immune system constitutes a cancer treatment. The novel therapy called the chimeric antigen receptor T (CAR-T) cell therapy for T-ALL is under investigation. The CAR-T cells targeting CD7, CD5, and CD1a have been developed to show therapeutic potential [18][19][20]. However, CAR-T-cell therapy remains a challenge for T-lineage neoplasms owing to its finite ability in distinguishing between the normal and malignant T cells; hence, it requires further development and trials [21]. So far, researchers have utilized ESCs as a vaccine for activating the immune system against cancer, but it has not shown much effectiveness and safety [22,23]. The clinical applications of ESCs have been obstructed mainly by ethical issues, tumorigenicity, and alloimmunization [22]. Therefore, searching for a new cancer vaccine to overcome these problems is necessary. The iPSC vaccines are undoubtedly a good tumor vaccine candidate to replace ESCs. Joseph et al. used the irradiated autologous iPSCs to produce iPSC vaccines inhibiting the recurrence of a variety of tumors and reducing the burden of metastatic tumors [12]. Tailoring iPSCs by introducing other driver mutations as a vaccine can prevent the initiation and progression of pancreatic cancer [24] and lung cancer [25]. Although the irradiated iPSCbased vaccines are effective at inducing immunity against multiple cancer types in mice, it is still fraught with risks such as tumorigenicity [26] and immunogenicity [27] which need to be overcome [28].
The iPSCs have been successfully derived from a variety of human cancers [29][30][31][32]. The differentiation of the leukemia-specific iPSCs is an effective strategy for studying the early stages of the normal and abnormal hematopoietic cells [33]. Kumano's laboratory has successfully constructed the iPSCs from the CML patients and then differentiated the iPSCs into blood cells sensitive to imatinib [34]. Besides, when the AML-iPSCs are differentiated into the blood cells, these cells regain the gene expression pattern of leukemia [35]. Muñoz-López and others also tried to generate iPSCs from the B-cell ALL patients [36]. Compared to the normal cell-derived iPSCs, the tumor-derived iPSCs may possess gene expression and epigenetic characteristics similar to that of the original tumor cells. Consistently, our results support the hypothesis that for T-ALL, the T-ALL-iPSC-based vaccines are more suitable than the H-iPSC-based vaccines.
Wu et al. demonstrated iPSCs alone to be incapable of stimulating an effective immune response and also stated that only the combination of iPSCs and adjuvant CpG can cause anti-tumor effects [12]. Our findings specified that the iPSC lysates could serve as complete antigens for stimulating the immature DCs and stimulating the killing effect of the T-ALL-iPSCs vaccine on the tumor cells. They are reportedly pulsed with the tumor cells or tumor cell lysates to obtain the vaccines and providing a variety of specific tumor antigens without prior knowledge of the targets [37]. Recent studies have reported that the fusion of the tumor cells with DCs can induce various anti-tumor immune responses [38]. The only non-antiviral cancer vaccine approved by the FDA is Sipuleucel-T (Provenge), a TAA-pulsed autologous dendritic cell-based prostate cancer vaccine [39]. The flow cytometry experiments showed that DC can enhance the function of the T cells, but produces no significant difference between the DC H-iPSC -T and Fig. 1 Pluripotency identification of the T-ALL-iPSCs. The pluripotency of the T-ALL-iPSC was identified by the alkaline phosphatase staining (A) and the expression of the pluripotent markers (B-D). Also, the karyotype analysis of the T-ALL-iPSCs was shown (E). When the T-ALL-iPSCs were transplanted into the immunodeficient mice, these cells could successfully form teratomas with three germ layers (F) ◂ the DC T-ALL -iPSC groups. This indicated that DCs had the same antigen presentation effect on the H-iPSC and T-ALL-iPSC.
This study clarified the inhibitory effect of T-ALL-iPSCs on the tumors and found that the DC H-iPSC -T and DC T-ALL-iPSC -T groups had similar cell-killing effects on the non-T-ALL tumor cells. Akin to the study by Nigel et al. the iPSC-based vaccine generated broad tumor immunity against various types of cancer. The iPSC vaccine generates a large number of tumor antigens for the immune system [12]. But co-culturing with the T-ALL cell lines including Jurkat, CCRF-CEM, and MOLT-4 showed that the DC T-ALL-iPSC -T cells significantly increased the cytotoxicity compared to the DC H-iPSC -T cells. These results indicated that the T-ALL-iPSC have a larger pool of T-ALL-related epitopes, and DC T-ALL-iPSC can induce the T cells to produce a specific killing effect against the T-ALL cell lines.
The patient-derived iPSCs were used as an antigen source for the anti-cancer vaccines because they can prime the immune system to respond to multiple patient-specific tumor antigens [40]. Hence, the patient's T-ALL cells were successfully reprogrammed into the T-ALL-iPSCs. When the H-iPSCs from the healthy people were compared to the T-ALL-iPSCs there were more similar gene expression profiles with the T-ALL cell lines, and the T-ALL-iPSC vaccine was found to stimulate more powerful toxic effects on the T-ALL cells. Therefore, T-ALL-iPSCs as a therapeutic tumor vaccine induces a better anti-tumor effect on the T-ALL. These beneficial properties make the iPSCs vaccine a potential choice for personalized adjuvant immunotherapy soon after the primary cancer treatment. Future studies should verify these effects through various in vivo experiments as they will establish the foundation for solid tumor cell-derived specific iPSC vaccine. Hence, Kooreman et al. describe a novel immunotherapeutic vaccination strategy using iPSCs that is highly effective in preventing multiple tumor growth in preclinical mouse models. Our results suggest that patient-derived iPSCs may have an advantage if targeted at a particular disease. Fig. 2 Analysis of the RNA sequencing data from the H-iPSCs, T-ALL-iPSCs, and T-ALL tumor cells. A The RNA-seq data of 9 H-iPSC clones generated from the healthy donors and T-ALL patients. The cell lines and iPSCs were subjected to a transcriptome expression correlation matrix, based on 15,687 filtered genes, which were generated. The healthy donor 1 (L1-iPS, L2-iPS, L3-iPS); healthy donor 2 (X1-iPS, X2-iPS, X3-iPS); T-ALL patient (W4-iPS, W6-iPS, W10-iPS); B based on more than 1000 over-expressed genes in the T-ALL patients, we were able to identify the presence of 61 genes that were highly expressed in the T-ALL-iPSC  The cytotoxicity (C), secretion of IFN-γ (F), and proportion of CD8 + CD107a cells (I) analysis of CTLs against the MOLT-4 cells were detected. Data were represented as mean ± SEM from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001