Tumorigenicity assay essential for facilitating safety studies of hiPSC-derived cardiomyocytes for clinical application

Transplantation of cardiomyocytes (CMs) derived from human induced pluripotent stem cells (hiPSC-CMs) is a promising treatment for heart failure, but residual undifferentiated hiPSCs and malignant transformed cells may lead to tumor formation. Here we describe a highly sensitive tumorigenicity assay for the detection of these cells in hiPSC-CMs. The soft agar colony formation assay and cell growth analysis were unable to detect malignantly transformed cells in hiPSC-CMs. There were no karyotypic abnormalities during hiPSCs subculture and differentiation. The hiPSC markers TRA1-60 and LIN28 showed the highest sensitivity for detecting undifferentiated hiPSCs among primary cardiomyocytes. Transplantation of hiPSC-CMs with a LIN28-positive fraction > 0.33% resulted in tumor formation in nude rats, whereas no tumors were formed when the fraction was < 0.1%. These findings suggested that combination of these in vitro and in vivo tumorigenecity assays can verify the safety of hiPSC-CMs for cell transplantation therapy.


Detection of malignantly transformed cells in vitro.
The soft agar colony formation assay and cell growth analysis are highly sensitive methods for detecting malignantly transformed cells in vitro 8,10,11 . In the present study, HeLa cells used as a positive control [11][12][13] . Colonies were formed by HeLa cells (1%) added to 1.0 × 10 4 human primary cardiomyocyte (HCM) after 10 days and even 0.5% HeLa cells on day 20. However, 201B7 cells, 201B7-CMs, 253G1, 253G1-CMs, 1231A3, 1231A3-CMs and HCM did not grow in soft agar ( Fig. 2A-C). We also analyzed the growth rate of human mesenchymal stem cells (hMSCs 1 × 10 6 ) contaminated with HeLa cells (10, 100, or 1000 cells) and hiPSC-CMs (Fig. 2D,E). The growth rate of hMSCs contaminated with HeLa cells has been reported as a control to detection of the immortalized/tumorigenic cells in human somatic cells 10 . These results indicated that the proliferation rate of HeLa cells contamination of hMSCs increased in dose dependent manner. However, the growth rate of hiPSC-CMs were constant and slow.
Karyotype analysis. We carried out a karyotype analysis in order to assess genetic alterations during hiPSC subculture and differentiation. It has been reported that the risk of aberrant hiPSC karyotypes increases with passage number; we therefore examined late-passage hiPSCs and hiPSC-CMs. There was no karyotypic aberrations in CMs derived from 20B7, 253G1 and 1231A3 cells during hiPSC subculture and differentiation (Fig. 4).

Discussion
Although hiPSC-CMs can potentially be used to treat severe heart failure, tumorigenicity limits their clinical application. Detecting and removing residual iPSCs or differentiated CMs that have undergone malignant transformation may be a key target to promise can ensure the safety of iPSC therapy. In this study, we established an assay for detection the potential tumorigenic cells in hiPSC-CMs in vitro and in vivo.
Although cervical cancer (HeLa) cells formed colonies in the soft agar colony formation assay, hiPSC-CMs did not exhibit colony formation. Moreover, cell growth analysis revealed that, HeLa cells proliferated notably faster compared with hiPSC-CMs having slow growth, therefore malignantly transformed cells were not detected in the in vitro assay of hiPSCs. TRA 1-60 and LIN28 are ideal markers for distinguishing residual undifferentiated hiPSCs among hiPSC-CMs by FACS and qRT-PCR. The latter was the more sensitive detection method of residual undifferentiated hiPSCs in hiPSCs-CMs. In the spike test, the detection limit was 0.001% by qRT-PCR as compared to 0.1% by FACS. In karyotype test, No karyotypic abnormalities were observed during hiPSC culture and cardiomyocyte differentiation. Additionally, in vivo tumorigenicity test, the mRNA expression of LIN28-a marker of undifferentiated hiPSCs-was associated with tumorigenesis in hiPSC-CMs transplanted into the heart of T cell-deficient nude rats.
There are a few possible mechanisms by which tumorigenesis can occur in iPSC-based therapy. In addition to teratomas, tumors may form from cells with tumorigenicity that have undergone malignant transformation during hiPSC culture and cardiomyocyte differentiation. The focus formation test, soft agar colony formation assay, and cell growth assay are widely used methods for detecting malignant transformed cells 8,10,11 . In the present study, we did not detect any malignantly transformed cells in the hiPSC-CMs using combination of soft agar colony formation assay and cell growth assay, providing one of evidences for the safety these cells.
Tumors can also arise from residual undifferentiated cells 14,[16][17][18] . A correlation between residual undifferentiated cells and teratoma formation after transplantation of neurospheres derived from mouse iPSCs has been reported 6 . It is necessary to use optimal stem cell markers for detecting residual undifferentiated cells depending on the differentiated cell type. Markers expressed also in target cells after differentiation cannot accurately detect residual undifferentiated cells in hiPSCs-derived products. Kuroda et al. have reported a method to detect residual undifferentiated cells in hiPSCs-derived retinal pigment epithelium cells (hiPSC-RPE cells) using Lin 28 in qRT-PCR 8 . LIN28 is involved in maintaining the self-renewal ability of stem cells, and it is one of the initialization factors used for inducing iPS cells from human somatic cells. As such, Lin 28 is highly expressed in human iPSCs. The results of our study strongly suggest that LIN28 is a marker that can be useful for detecting undifferentiated residual cells in hiPSC-CMs by qRT-PCR, as it showed the highest sensitivity of all the Stem cell markers tested, with no expression in terminally differentiated cardiomyocytes and highly expression in hiPSCs. In addition to residual undifferentiated cells or malignant transformed cells, critical genomic changes and the survival of foreign genes in iPS cells have some possibilities for tumor formation in hiPSCs-derived products. The karyotype analysis, which has capability in detection of extremely large genetic abnormalities, revealed no major genomic changes in hiPSC-CMs. To check genetic abnormality in detail a SNP array et al., which can analyze small or large genomic variations in whole genome sequence with extremely high sensitivity, may be critical for confirming the safety of hiPSCs-derived products for therapeutic applications. However, the correlation between genomic abnormalities and tumorigenicity remains to be determined. A more detailed analysis of these would require further study. The results of our study demonstrate that hiPSCs-derived products with high LIN28 mRNA expression exhibited a higher rate of tumorigenesis, indicating that LIN 28 can predict tumorigenesis after transplantation of hiPSC-CM. However, this must be confirmed in different type of immunodeficiency animal model, routes of  tumorigenicity assays are costly and time-consuming. Moreover, some degree of skill is required to transplant cells into rat or mouse heart. We suggest that in vitro assays which detect the malignant transformed cells and LIN28 expression level may be substituted for in vivo assays.
In conclusion, we developed an assay that combines in vitro quantification of tumorigenic cells and in vivo tumorigenicity assessment to verify the safety of hiPSC-derived CMs for regenerative therapy of heart failure or heart disease. Further studies are warranted to verified whether this in vitro system can ensured the safety of hiPSC therapy for the clinical application of cell transplantation therapy using human iPSC-CMs.

experimental procedures
Animal experiments were performed according to the Guide for the "Care and Use of Laboratory Animals" (National Institutes of Health publication). Experimental protocols were approved by the Ethics Review Committee for Animal Experimentation of Osaka University Graduate School of Medicine (reference number; 25-025-034).

Cardiomyocyte differentiation and purification. Cardiomyocyte differentiation and purification.
Cardiomyogenic differentiation of 201B7 and 253G1 cells was induced as previously described 19,20 . To differentiate 1231A3 cells into the cardiac lineage, we used a modified version of an established protocol 21 . Briefly, embryoid bodies (EBs) were generated in an EZ sphere (Iwaki, Shizuoka, Japan) in Stem Fit Ak03N without basic fibroblast growth factor (bFGF; R&D Systems, Minneapolis, MN, USA) in the presence of 10 μM Y-27632 (Rho-associated protein kinase inhibitor; Wako Pure Chemical Industries, Osaka, Japan) and bone morphogenetic protein (BMP)-4 (R&D Systems). After EB formation, the culture medium was replaced with differentiation medium containing Stem Fit Ak03N without bFGF and supplemented with several human recombinant proteins including BMP-4 and activin A, bFGF, and vascular endothelial growth factor (all from R&D Systems) as well as the small molecule inhibitor of Wnt production 3 (Stemgent, Lexington, MA, USA), SB431542 (Sigma-Aldrich, St. Louis, MO, USA) and Dorsomorphin (Sigma-Aldrich). After cardiac differentiation, the cells were dissociated with 0.25% trypsin/EDTA (Thermo Fisher Scientific, Waltham, MA, USA), cells were plated onto the gelatin coated dish. HiPSC-CMs were maintained in Dulbecco's modified Eagle's medium (DMEM; Nacalai Tesque, Kyoto, Japan) containing 10% fetal bovine serum (FBS; Sigma-Aldrich) or DMEM without methionine (Thermo Fisher Scientific) for eliminate residual undifferentiated cells 22  RNA extraction and qRt-pCR. Total RNA was extracted from hiPSCs and hiPSC-CMs using the PureLink RNA Mini kit (Thermo Fisher Scientific) and used to synthesize cDNA with the Super Script III First-strand Synthesis System and random primers (Thermo Fisher Scientific). qRT-PCR was performed on a ViiA 7 RT-PCR system (Thermo Fisher Scientific) using SYBR Green and human stem cellspecific primers 8 (Table 1). Each sample was analyzed in triplicate for each gene studied. The average copy number of gene transcripts was measured relative to the level of glyceraldehyde 3-phosphate dehydrogenase in each sample.
Immunofluorescence analysis. HiPSC-CMs were fixed with 4% paraformaldehyde and labeled with antibodies against cTNT (Abcam, Cambridge, UK) and anti-sarcomeric alpha actinin (α-actinin (Sigma-Aldrich) as primary antibodies, followed by secondary antibodies such as Alexa Fluor 488-or AlexaFluor555 conjugated donkey anti-rabbit or -mouse IgG (Thermo Fisher Scientific)antibodies. Nuclei were counterstained with Hoechst 33342 (Dojindo, Kumamoto, Japan). Images of samples were acquired on a confocal microscope (Fluoview FV10i; Olympus, Tokyo, Japan). soft agar colony formation assay. The soft agar colony formation assay was performed using the CytoSelect 96-well cell transformation assay kit (Cell Biolabs, San Diego, CA, USA). Human cervical carcinoma cell line HeLa cells (JCRB Cell Bank, NIBIOHN, Osaka, Japan) were used as a positive control since they are known to form colonies in soft agar medium (Kusakawa et al. 11 ; Ke et al. 12 ; Seo et al. 13 ). The cells were maintained in Eagle's minimum essential medium (Sigma-Aldrich) supplemented with 10% FBS (Sigma-Aldrich) and 0.1 mM non-essential amino acids (Thermo Fisher Scientific). The primary human cardiomyocytes (HCM) (PromoCell, Heidelberg, Germany) were were maintained in Myocyte Growth Medium (PromoCell). HeLa cells Cell growth assay. Cell growth assay of hiPSCs was evaluateld according to Kono et al. 10 . Human mesenchymal stem cells (hMSCs) (Lonza, Walkersville, MD, USA) were maintained in mesenchymal stem cell growth medium bullet kit (Lonza). A total of 1 × 10 6 hMSCs were mixed with 1000, 100, or 10 of HeLa cells and used as a positive control. HiPSCs-CMs were dissociated with TrypLE Select (Thermo Fisher Scientific) and 1 × 10 6 hiPSC-CMs were seeded in T175 flasks (Corning Inc., Corning, NY, USA) and cultured. This process was repeated for 10 passages and growth rate was calculated.

Karyotype analysis.
Thymidine was added to hPSCs-CMs cultured at 37 °C under 5% CO 2 and cultures were synchronized from 10 to 24 h. After washing to remove the thymidine, colcemid was added and culture synchronization was carried out for 6 to 24 h to obtain metaphase cells, which were fixed and added as a drop onto a slide glass to prepare chromosome specimens. The preparation was stained with Hoechst 33258 and quinacrine mustard solution to obtain Q bands. Chromosomes were classified according to band pattern and karyograms were prepared. A karyotype of up to 20 cells was analyzed in the chromosome karyogram.
HipsC-CM sheet transplantation into nude rats. Male F344/NJclrnu/rnu rats (5-6 weeks old, weighing 190-250 g) were obtained from CLEA Japan (Tokyo, Japan). A total of 1 × 10 7 cells per hiPSC-CMs sheet was transplanted into the left ventricular surface of each rat heart. Rats were sacrificed 4 months later. All surgeries and sacrifices were performed on deeply anesthetized animals.